Embed Size (px)
Citation preview
C3
TECHNICAL REPORT GL-83-3
CONCRETE BLOCK" -PAVEMENTSby
Raymond S. Rollings
Geotechnical LaboratoryU. S. Army Engineer Waterways Experiment Station
P. 0. Box 631, Vicksburg, Miss. 39180
JUT 0ICMarch 1983NFinal Report
Approved For Public Release; Distribution Unlimited * ~ --...
j~r 4IRV
rprdfor Office, Chief of Engineers, U. S. ArmyWashington, D. C. 20314
Project No. 4AI11O12AT22, Task Area AC,V K Unit 005
Destroy this report when no longer needed. Do not returnit to the originator.
The findings in this report are not to be construed as an officialDepartment of the Army position unless so designated.
by other authorized documents.
The contents of this report are not to be used foradvertising, publication, or promotional purposes.Citation of trade names does not constitute anofficial endorsement or approval of the use of
such commercial products.
UnclassifiedSECURITY CLASSIFICATION OF THIS PAGE (ndn~ Data Enter*Cd)
REPOT DCUMNTATON AGEREAD INSTRUCTIONSREPOT DWAINTATON AGEBEFORE COMPLETrING FORMA
1REPORT NUMBER 2.GOVT ACCESSION No. 3. RECIPIENT'S CATALOG NUMBER
Technical Report GL-83-3 'qi 14. TITLE (andSubtfl) 5. TYPE OF REPORT A PERIOD COVERED
CONCRETE BLOCK PAVEMENTS Fnlrpr
6. PERFORMING ORG. REPORT NUMBER
7. AUTHOR(*) S. CONTRACT OR GRANT NUU8ER(a)
Raymond S. Railings
9. PRrORMING ORGANIZATION NAME AND ADDRESS III. PROGRAM ELEMENT. PROJECT. TASK(
U.S. Army Engineer Waterways Experiment Station AREA & WORK~ UNIT NUMBERS
Geotchnial LboraoryProject No. 4A161102AT22,P. 0. Box 631, Vicksburg, Miss. 39180 Task Area AO, Work Unit 005
MCONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE
March 1983Office, Chief of Engineers, U. S. Army ~ 4MEQ AE
Washington, D. C. 2031413
14. MO0NIYORING AGENCY NAME 4ADDRESS(to dilloe"a 11M COM,41Uj" OffI)Ce) is. SECURITY CLAWS (O1 two 'OeOIU
Unclassified
SCMHtCUL
I Approved for public release; distribution unlimited. -
I?.WSR~4IIO 5NOMET(1I.hc i4d.*~&2.I U~e o
Available from National Technical lufornutiou Service, 5285 Port Royal Road,Springfield, Va. 22151.
Conicrete blocks
Rigid pavemenitsTra~ffic testa
describes an aceleated trafic test of the pving black test item con- Aducted at the U. S. AryEngineer Vaterw~ys Euperiment Station, This intor-tion is i ,xy_,ed to develop recomendations otu Rpecifications aud drsignsethods for block paveats, for use by the U. S. Aray Co&Vs of E£ueers.
(continued)
W U13 mUvlow o 'swo k& Oevess
Unclasifie
iII
UnclassifiedSECURITY CLASSIICATION OF THIS PAOC(-a "a fti4re.
20. ABSTRACT (Continued).
S - ' In general concrete block pavements behave as flexible pavements, and .designs may be developed from modifications to existing flexible pavement de-sign methods. Because of the superior load distributing characteristics ofblock pavements in comparison to conventional asphaltic concrete surfaces,these designs will be conservative. Both rectangular and shaped, interlockingblocks may be used in pavements. Concrete block pavements are acceptable aslow speed road and industrial surfacings, are capable of supporting heavy andabrasive loads, require little maintenance, and offer easy subsurface accessfor utility repair or corrections for settlement.
<I "
I
, ! I
I '{.. ," " ~l~lst~krli -V
tI
I.• ., .
• .. . ..... .-1, .. . .. . . .. .. .,... ., ,.. 1 .. i ::, :..,.. ::: C...r
PREFACE I
This investigation was conducted by the Geotechnical Laboratory
period November 1978-September 1980. The study was aponsored by the
Office, Chief of Engineers, U. S. Army, udrProject No. 4A161102AT22,
Task Area AO, Work Unit 005, "Analysis of Precast Articulated Pavement
System Units."
This study was conducted under the general supervision of
Mr. J. P. Sale, former Chief, GL, Dr. W. F. Harcuson 111, Chief, GL,
Mr. A. H. Joseph, former Chief, Pavement Systems Division-(PSD), GL,A
and Dr. T. D. White, Chief, PSD. The study was conducted by Mr'. R. S.
Rollings, PSD.
Commanders and Directors of WES during this investigation and
jpreparation and publication of this report were COL John L. Cautnon, CE,'I COL Nelson P. Conover, CE, and COL Tilford C. Creel, CE. Technical
Director was Mr. F. R. Brown.
A
Fr
*t t
01s I pci?
CONTENTS
Page
-PREFACE. .. ... ....... ....... ....... ..... 1
CONVERSION FACTORS, U. S. CUSTOMARY TO METRIC (SI)UNITS OF MEASUREMENT .. .... ....... ....... .... 4
PART I: INTRODUCTION .. .. ...... ....... ........ 5
Background. .. ....... ....... ........... 5Scope ..... ... ...... ....... .... ...... 5Description .. .. ...... ....... .......... 5
-Historical Developm~ent. .. ....... ....... .... 8Applications .. .. ... ....... ....... ..... 10
PART II: MANUFACTURE OF PAVING BLOCKS. .. ....... ..... 12F
Equipmuent ... .. ... ....... ....... ....... 12Concrete Mixture Proportioning...............13Block Quality. .. .... ...... ....... ..... 16
PART III: BLOCK PAVEMENT CONSTRUCTION.. ....... ..... 24
PART IV: PERFORMANCE. .. ... ...... . . ... .. .. ... 30
Roads and Streets .. .. ...... ..... .. .. ... 30Industrial Applications .. .. ...... ....... ..Port Facilities... ...... ....... ........ 34
F ~Drainage. .. ....... ............ 36[Cost and Hiistenance. .. ...... ....... ..... 38Surface Proper'ties .. ... ...... .. .. .. .. . .41
PART V; LOCK PAVEMIENT TESTS..... .. .. .. ... 45
Unted KingdomTests . .. ....... ......... 45 .-
New Zealantd-Tents, Canterbury Circ~ular Test Track . .52
Australian Tests.............. 55South African~ Tests, H~eav~y Vehicle simulator ..... 63
D anish TestRoad . . . . .... . . 67
IPART VI: WES TEST SEC1rION .. . . . . . . . . . . . . . . 70
Objective.......................70V ~Tet Secon esripii0,11 .... . .... 70
..................................... .. .. .. ..ITraffic. ; .. .. .. .. .. .. ... .. . . . . .79 -
I~~ Peftuc. ..... ......... . . . . . . . 83Analysis of Resu....................... . . ....N. 0[I Con~clusions...................**.
I.j.
Page
PART VII: ANALYSIS AND DESIGN. .. .... ....... ..... 106
Behavior. .. .... ....... ....... ....... 106 -
Block Shape and Laying Pattern .. .. ....... ..... 107 -
Block Thickness. .. .. ...... ....... ...... 108Laying Course. .. .. ...... ....... ....... 109Design .. .. ...... ............ .. .. ...... 109Block Strength .. .. ...... ....... ........ 113
PART VIII: CONCLUSIONS AND RECOMMENDATIONS .. .. .... ..... 115
Conclusions .. .. .... ....... ....... .... 115Recommendations. .. .. ...... ....... ...... 115
REFERENCES. .. .... ..... .. .... ........ .... 117
TABLES 1-20
J
3j
1A
CONVERSION FACTORS, U. S. CUSTOMARY TO METRIC (SI)UNITS OF MEASURE MENT
U. S. customary units of measurement can be converted to metric
(SI) units as follows:
Multiply By To Obtain
cubic yards 0.7645549 cubic metres
Fahrenheit degrees 5/9 Celsius degrees or Kelvins*
feet 0.3048 metres
inches 2.54 centimetres
kips (force) 4.448222 kilonewtons
miles per hour 1.609347 kilometres per hour(U. S. statute)
pounds (force) 4.448222 newtons
pounds (force) per 157.0874585 newtons per cubic metrecubic foot
pounds (force) per 0.271447 newtons per cubic metrecubic inch
pounds (force) per 5.8180544 newtons per cubic metrecubic yard
pounds (force) per 47.88026 pascals
square foot
pounds (force) per 6.894757 kilopascalssquare inch
square feet 0.09290304 square Wetres
square inches 6.4516 square cenitietres
square yards 0.8361274 square metres
tons (force) $"96.444 newtons
" ! I
* To obtain Celsius (C) temperature readings from Fahrenheit (F) read-ingst use the following formula: C = (5/9)(F - 32) , To obtainKelvin (K) teadiags, use: K (5/9)(F - 32) + 273.15
41,.........
• ' , . ..- .. . , , , , , .. . . .• , . ., .. . .. . . .V,. . :,
-p :,, . : .. . . : , ,: ' . : . , ' . '.:. ,::: : . ,.,,. .,. , : - ,,,.i:. : ,,, !: :.?.i .
CONCRETE BLOCK PAVEMENTS
PART I: INTRODUCTION
Background
1. Concrete block pavements are an established pavement surfacing
that competes successfully with conventional portland cement concrete
and asphaltic concrete in Europe for many uses. In the United States
the concrete paving block industry is relatively new but is growing.
The U. S. Army Corps of Engineers (CE) has used paving block in Europe
and on at least one project in Florida in the United States. Use of
block paving in the United States and in Corps projects may increase in
the future, but there is little infiatuation available to the CE on the
design, construction, and performance of block pavements.
2. This report sumarlzes available information oi solid, con-
crete block pavement, describes several installations of paving block,
and reports the results of an accelerated traffic test of block pavetuent
conducted by the Geotechnical Laboratory of the U. S. Army Engineer
IWaterways Experioent Statioi (WES). This information is used to recoM-
mead design procedures, specifications, and areas for further work with -
block pavements.
1t
3. A block pavement consists of euse, accurately ditwnsioned con- A
.- crete blocks Which fit closely together to form a pavement surface. The
- *§. blocks arc oaufactured in a wide variety of shapes, softe of which are jshown iu Figure 1. Generally the blocks are about the size of a coaon.
- -. brick with a thickness of 2-3/8 to 4 intS, and weigh about 9 to 12 lb
* A table of factors for converting U. S. customary units of measure- T
sment to metric (SI) units is presented-on page 4.
.tI
Figure 1. Examples of paving block shapes
each. A thin, 1- to 2-in.-thick leveling course of sand is used underthe blocks. The blocks are generally laid by hand on this sand layer.The blocks are then compacted with a manually operated vibratory platecompactor which seats the blocks in the sand layer, compacts the sandlayer, and forces some sand into the joints between blocks. Additionalsand is then swept into the joints. between the blocks, and move passesare made with the vibratory plate compactor to compact and wedge thissand into the joints. A base and subbase course under the levelingcourse provide s ructural support similar to that of a conventionalflexible pavemeat. Figure 2 shows a generalized cross section of ablock paveaent. Hany different patterns of laying blocks aie. possible,and several pAttecus are illustrated in Figure 3.
4. The solid concrete paving blocks are also commonly calledpavers," "inteorlocking paving stone," "road stones," or ".Interlockig-
paving- block," The different shaips of paviog block are often ideuti-.ied by manufacturers' trade ams such as "!aistou Te," "Finetta, 9 "
Paver," etc.5. Several manufLcturers also produce coucrete grid paving blocks.
These grid paving blocks are generally 16 by 24 in. or 24 by 24 in.u and.4 in. deep with vrious sizes, pattensi and shal~s of openings in thei .i."iii surface. The open spaces in the stve are filled with top soil, seeded, .
. grass par y c s .. bk ee
* 1 " : '" : " '
CONCRETE BLOCK SURFACE
.. .'.SAN LEVss ELtI o lckpv~
BAE OUS
0.SkodI oa i ~wubi
FigtiKe 3. Crosps secio ofa lockn~ pvemen
AI
7 *
for light traffic. Both solid and grid concrete paving block have also
been used for erosion control. This study considers only solid concrete
paving block used for pavements. Low strength concrete blocks (patio
blocks, etc.) intended only for pedestrian traffic will not be covered
in this report, In 1979, nine city blocks of Chicago's State Street
were repaved for pedestrian traffic with hexagonal asphalt blocks
(Asphalt Institute 1979), and similar asphalt blocks were used for pav-
ing an open area at the University of Maryland. Asphalt paving blocks
will not be covered in this report.
Historical Development
7 6. Stone blocks, bricks, cobbles, and composite wood and tar
units were used for road surfacings up to World War I. After this time,
these paving units largely disappeared due to increased construction
costs, surface smoothness requirements, and the availability of more eco-
Suooical, less labor intensive alternatives. Manufacturing technology in
the 1950's allowed mass production of accurately dimensioned, high-
strength concrete blocks and several paving block designs were intro-
duced ii Europe during this time. Europe has a long history ri paving
with individual stone blocks, and the concrete block pavements were
* readily accepted i .continental Europe.
* 7. Small modular paving elements have been used in Amsterdam
* since the Hiddle Ages with natural stone in the earriageways to cesist . .
abrasion f to steel wheels and sleds and bricks it pedestrian areas
(Kellerann 1980). As rubber tires becae comon at the end of the
nineteenth century, bricks replaced the morQ mientive stone in the
* 1 cattigeays. The Waalformat bricks 7.7 by 3.3 by 1.9 in. (ISS by 85 by
48 ao) became the aost caoon paving utit, but as the character ot' ve'
bicular traffic changed, the thickness grew progressively from 1.9 in.
.(48 m) to 2.5 in (64 ow) and then 3.6 in. (92 o),. After World Var II
V .. the supply of traditional bricks failed -to keep up with demand aud by
1955 only half of the deand for paving bricks could be met (Kellersmaun
1980). In 1951 the concrete prodtwte company Uollad began production
4S
"0w "' . .. .4- ,.-" , ,g
of a rectangular concrete paving block and was followed in 1952 by the
'I Schokbeton Company with an I-shaped concrete paving block (Van der Vlist
1980). The concrete paving block was readily accepted as a substitute
for the scarce paving brick and today has essentially replaced it due to
much lower costs. Since 1960 the Netherlands paving block industry has
been highly mechanized and automated, and as can be seen in Figure 4,
its growth.has been steady (Van der Vlist 1980).
20
15
0<0. n
10
1'0
978 960 1 62 63 64 65 66 67 68 6970 71 72 73 74 7678 7778 79
Figure 4. Concrete paving block production in theNetherlands (Van der Vlist 1980)
8. The use of concrete paving block in the Netherlands developedI. naturally from existing pavement practices and construction procedures.
Consequently, it has never been perceived as a novel or new product and
has always been readily accepLed. To a somewhat lesser extent this was
true throughout continental Europe, and the concrete paving block indus-
try has developed strongly in this area, particularly in the Netherlands, -.
the Federal Republic of Germany, and Denmark.
9. The first British concrete paving blocks were produced in
1968 under a German license but were generally limited to architectural
roles. The United Kingdom (UK) Cement and Concrete Association (CCA). condacted a test and evaluation-program of paving block agd studied he .
continental Europe~n paving block industry to try: to widen paviklg block
f evlJinPormoavn lc n-tde
.,!J . -':)~.. , flf.i -*
... . . . . . . . . . . .. . . . . . . . . . . .. . . .. . .. . .
use in the U.K. (Knapton and Barber 1980, Knapton and Lilley 1975).
A large promotion and education program in 1976 successfully increased
the use of paving block in road and industrial construction in the
United Kingdom. South Africa, Australia, and Canada also have growing
concrete block paving industries.
10. The United States has used brick, stone, and wood for road
surfacings since colonial times, and many of these old road surfacings
still exist in historic districts of many cities. Bricks were used as
a pavement surfacing as early as 1832 and remained competitive until
after World War I (Wiley 1919). They were gradually displaced by port-
land cement and asphaltic concrete pavements. Concrete paving blocks
were first produced in the United States in the 1960's using German
equipment and designs. Today there are a number of manufacturers pro-
ducing paving block, and trade associations, such as the National Con-
crete Masonry Association (NCMA) and Interlocking Paving Manufacturers
Association (IPHA), are involved in promotional and educational programs
for concrete paving block. The American Society for Testing and Hate-rials (ASTM) is also preparing a standard specification for concretepaving blocks, but concrete paving blocks are still widely considered
as a new paving material in the U. S.
Applications
11. The aesthetic value of concrete paving blocks is generally
recognized, but they also provide a high-strength surface which is re-
sistant to environmental damage and is capable of supporting large con-centrated loads and heavy traffic under abrasive conditions and which is
resistant to environmental damage. Because of the modular structure,- ... blocks can be removed from the surface to allow acceas to subsurface
utilities, or to correct settlement of underlying material. Ninety toninety-five percent of the original blocks can then be used to resurface
. ' the pavement.12. block pavements have higher initial costs than conventional
pavements. The same cost relation may not hold for life-cycle costsf>*~ 10' '.4i 4 j0
. . "-" .,., . ,
that include the reduced maintenance costs of concrete paving blocks.
Not enough data are currently available to adequately evaluate life-
cycle costs of block pavements in the United States. The surface of a
block pavement is rougher than conventional pavements, and this effec-
tively limits maximum vehicular speeds to less than 40 mph.
13. In L"rope paving blocks find their major use in low-speed
road pavements and industrial applications. In 1972, over two-thirds
of West Germany's production of 30 million yd of paving block produc-
tion was used in road pavements and industrial applications. In the
same year Denmark used 40 percent of its paving block production for
industrial construction (Cement and Concrete Association 1976). Table Ishows a detailed breakdown on the use of the Federal Republic of 1
Germany's block production.14, In the United States, the paving block market is still devel-
oping. Individual manufacturers emphasize different marketing targets.
Some are concentrating on the aesthetic market, such as driveways, pool
.!decks, and parking areas in expensive developments. At least one manu-
facturer is trying to develop a home owner "do-it-yourself" market.
Other manufacturers are emphasizing industrial and municipal road appli-cations, such as steel mills and resurfacing municipal roads.
I! "i j
i . ii,
1.. .11• ".
- - 4 + - ,
- !I
PART II: MANUFACTURE OF PAVING BLOCKS
Equipment
15. Concrete block production is highly automated to mass produce
the product economically. Two basic types of equipment are used to pro-
duce concrete paving blocks today: the conventional block machine and
the multilayer machine.
16. The block machine is used to produce conventional concrete
masonry block, but with minor modifications it can also produce concrete
paving block. In this process, a dry stiff concrete is forced into
molds under pressure and vibrated intensely at high frequency. The com-
pleted block then leaves the machine for initial curing. Steam curing
is sometimes used. This reduces the ultimate strength of the paving
block, but increases the early strength of the paving block to allow
handling. After initial curing to gain strength for handling, the
blocks are placed on pallets and stored while curing continues. Produc-
ing paving blocks on a conventional block machine allows a block manu-
facturer to supplement his conventional concrete tuasonry block produc-
tion with paving blocks during periods of low masonry block demand. In
the U. S. some block manufacturers have expanded into pay ;tg block pro-
duction to keep their equipment in use.
17. The multilayer machines are specifically designed for mass
production of concrete paving block. As before, a dry, stiff concrete
mix is forced Into molds under pressure and subjected to intense vibra-
tion. An entire array of blocks sufficient for one layer on a shipping
pallet is cast at one time, After the mold is removed, a thin layer of
1:7 sand is spread over the newly cast paving blocks, and then another layer
of paving block is cast directly on the sand-covered lower layer. Thiscontinues until an entire pallet of approximately 8 to 10 layerE of pav-
ing block is cast. If a stationary multilayer machine is being used,
the pallet of paving block leaves the machine, and another pallet of
blocks is cast. If a traveling multilayer machine is used, the pallet
remains in place, and the machine moves on rails and casts the next
12
-. . . . . ,,........ ...... . . . .. ,,,"
A :
* I pallet of paving block adjacent to the first pallet. These paving
blocks are generally air-cured and may be covered with polyethylene to
prevent drying during curing. Figure 5 shows a completed polyethylene
covered pallet of Z-shaped paving block produced by a multilayer machine.
J, "I.
Figue S.Finshedpalet o paing loc
Figh cnrtmxure sed palo paving block ealyijr
wi 8. Ptr-oucteet a ocu les to mol ad weally ipoetainsro
potonng iecnqte Ibcematonal m xmpe ogdeecie pd- n
ucpavie sown i grell 6/.wihartofabu 0pcetsn
pedooncialemtretes macinerqieadtxture andotonn qui
19. ofthe n t itreue o paving block, OUS.mnfcursggeslyts dy
.v13
with ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~4 aAe-ocmn ai fls ta . n omlycnan
-W 17 MM . .....
W., I 4v
F~.gre . Dfecive avig bockiniia tia mi wthanaggegtefienss odlu o 3.0 o .6 a
a~~~N. satnponfoprpringthmi(eseCopn(ndAte)Dawson~~~~~~~~~~~~~~~' (18)poie oedtie icsino rotnigo-
Al crete mixe~~Srainblcs
kowsk 1980). Figre 6.sefecv paving blocks esnhtco auainide tat mixide wnaegaen stables ols uc 3.s tod yelloas
bDawson, a980 priish acr.alddicsinorootoigca
21. mixe or ainteocst.fpvn lc anb nrae pt
T 2is Poimons ian be adedote 7.ncrte optinguloc pmitotn
povbighe syntloetipocths pigments sa aou inerct oer theneaelt
paing h o iv (Von Szadkowski 1980).Alhuhogncp-
kok 228. olrPigent ise inr pavn plocs oar iynthei crhnaicalfro oxdst14rvd eahrn tbeclssc srd elwbuf brwn an ryshbak
21.- Th oo nest fpvn bokcnb.nrae pt
V - - - _ _ . - . - - - - . - - - . - . . ~ - - . - . . . . -
6011UFSLU
0 OPTIMUM AODITION
0 1 2 3 4 5 8 7 a 10
PIGMENT, PERCENT
Figure 7. Effect of pigment content on colorintensity (Dawson 1980)
reactions of the concrete. If the water-cement ratio of the concrete
mix is kept constant, the addition of pigment will not affect the final
product strength. However, as shown by the spread test in Figure 8, theworkability of the mix may dcucreasie when the pigment is added at a
NO PIGMENT 10% BLACK'
8IN. 8. IN.
10% BUF10% lIeD
F~igure 8. Effect of pigmaent content on spread at aconstant water content (Dawson 1900)
j .[-4
iif
constant water content, and this may lead to manufacturing problems such
as adhesion to the manufacturing machine's rams. To restore workability,
I j additional water may have to be added to the mix and this additional
water may lead to a reduction in the block's strength.
BlockQuality
23. Concrete paving blocks must have sufficient strength and
durability to withstand traffic loads, abrasion, and weather conditions.
They also must be manufactured to close dimensional tolerances to allow
rapid construction with tight joint patterns. Various organizations
have specified different test standards to ensure that paving blocks
will have the required properties to perform their function, but these
standards show some variation.
Strength
24. The most common requirement for blocks is a minimum strength.
Table 2 compares several different paving block strength requirements
from different countries. These strength requirements are high, but
blocks with these strengths have an established history of good perform-
ance under severe loads. The consistent high quality of paving blocks
is believed to be one reason for its rapid acceptance and growth (Kuthe
1980).
25. Theie is considerable disagreement over the appropriate
strength test that should be used to evaluate paving blocks. Various
proponents have -suggested compressive, flexural, and splitting tensile
tests. Paving block strength requirements will be discussed in moredetail in Part VIII of this report.:- .-
Abrasion .. : .
26. A block pavement is subject to abrasive wear from traffic anda var'iety of abrasion tests are possible. Tire Netherland's standard
NEN 7000 requie'es a sandblast abrasion test to evaluate abrasion resis-
tance. Dreijer (1980) describes the abrasion-related problems which
developed iW 'the Netherlands in 1968 and the resulting investigation and
V.. changes in theNN 7000.standard that ve made to avoid theseproble
in the future. This investigation found a direct relationship between
the weight loss in the sandblast test and the block flexural strength,
but blocks with a thin abrasion vulnerable surface layer could not be
identified by strength alone. This thin layer generally developed from
curing techniques, improper plasticity of the concrete mix, or decora-
tive finishing. The sandblast test was effective in identifying these
thin abrasion-susceptible layers.
27. The German standard DIN 18501 on paving blocks issued in 1964
included a requirement for a mechanical grinding test to evaluate abra-
sion resistance. A proposed revision to DIN 18501 will drop this re-
quirement because experience has shown that blocks with the required com-
pressive strength of 8700 psi are sufficiently abrasion resistant (Meyer
1980).
28. There are a variety of abrasion tests available for concrete
such as sandblast tests, rattler-type tests, or mechanical abrasion
tests with disks, wheels, or steel balls. However, these tests only al-
low an evaluation of relative quality without any defined acceptable
criteria for wear of concrete surfaces (Lane 1978). Concrete abrasion
resistance is affected by a variety of factors, but concrete compressive
strength has been widely used as an abrasion criterion and is used to
set abrasion resistance standards for industrial floors (American Con-
crete Institute 1969, Spears 1978). The high compressive strength ofconcrete paving8 block indicates high abrasion resistance, but evaluatioa
only on the basis of strength will not identify the thin abrasion-
susceptible block surfaces reported by Dreijer (1980).
29. An abrasion test appears to be needed in paving block speci-
fications to identify blocks with abrasion-susceptible surfaces. The
selected abrasion test may have to be modified. For example, Dreijer
(1980) reports that the amount of sand used in the sandblast test was
reduced from 3500 g to 1000 g so that only a thin surface layer was
actually tested and not the underlying concrete. The abrasion resis-
tance of the lower concrete could be evaluated adequately on the basis
, of tbe block strength.
17' . y.,, ,
. ... . .
. + " . . . . ' -:+. . ... , .- , -: A
Freezing and thawing
30. Deterioration can occur to nonfrost-resistant concrete when
it is subject to critical saturation with water followed by freezing and
thawing. If water freezes in concrete pores that are large enough to
contain freezable water and that are critically filled, the expansion of
the ice will try to expel water from the pore. Depending on the speed
of freezing and the permeability of the cement paste, dilation pressures
can develop (Neville 1973, Mather 1975, Powers 1975). Pores large enough
to contain freezable water can exist in both the concrete aggregates and
the cement paste. Another source' of dilating pressure which can cause
freezing-related concrete deterioration is the osmotic or osmoticlike
pressure developed from local increases in solute concentration due to
the separation of frozen water from the solution (Neville 1973, Powers
1975). This mechanism may be particularly important in concrete pave-
ments. A concrete pavement slab which freezes from the top can be seri-
J ously damaged if water has access to the bottom of the slab and travels
through the slab due to osmotic pressure (Neville 1973). The concrete
moisture content can increase above its original value and segregation
of ice crystals into layers has reportedly been observed in some cases
(Neville 1973). The use of deicing salts is believed to further in-
crease the solute concentration near the slab surface with a resulting
increase in the osmotic pressure (Powers 1975). Paving block surfaces
may be exposed to environmental and salting conditions similar to that
of conventional concrete pavements.~~31. lThe major factors that determine the resistance of' concrete
to freezing and thawing are the degree of saturation and the pore struc-
ture of the concrete (Neville 1973). Concrete can be protected against.
freezing by providing a properly air-entrained paste and ucing sound
frost-resistaut aggregate (Neville 1973, Hather 1975, Powers 1975). All
paving blocks should contain sound frost-resistant aggregate; however,
air entraitnmnt is not now used in concrete paving block. The intense
vibration used in paving block manufacture is claimed to cause an unde-
sirable loss of entrained air, and Clark (1980) states that the stiff
I - •onsistency of the low water-cement ratio mitures used in paving blocks
18
" - --.-- " ._
• . ? •.. .
,. . ~ , .. .. , • . . . .. . . . . , = "
inhibits the action of air-entraining agents and makes measuremeet of
the air content very difficult. Therefore, the practice in Europe has
been to specify a paving block strength that is hoped will provide pro-
tection against frost and deicing salts and not to use entrained air.
32. Meyer (1980) states that paving blocks that meet the 8700-psi
compressive strength requirement of the German DIN 18501 and are manu-
factured from standard cements and frost-resistant aggregates are suffi-
ciently durable when exposed to freezing and thawing and deicing salts.
Dreijer (1980) reports that the blocks meeting the strength and sand-
blast test requirements of Netherlands NEN 7000 can be expected to be
resistant to frost and deicing salts. If paving block will not be ex-
posed to freezing and thawing, the United States National Concrete Ma-
sonry Association (1979) reduces the required strength from 8000 psi to6000 psi and increases the maximum allowable absorption from 5 to 8 per-
cent. However, they require that the durability of the paving blocks
that will be exposed to freezing and thawing be established by proven
j.1 field performance under similar field conditions for 3 years or by con-
ducting a laboratory freeze-thaw test.
33. Clark (1980) describes a series of tests of concrete paving
blocks that examine the effects of density, water-cement ratio, cement j .content, compressive strength, 24-hr absorption values, and initial sur-
face absorption on resistance of paving blocks to freezing and thawingwith deicing salts. The test specimen paving blocks were prepared with
various types of aggregate, with ceent contents varying from 318 lb/yd3 . .-
to 989 lb/yd3, and with water-cement ratios of 0.22 to 0.62. Test spec- .
iens were also made from paving quality concrete with 3 to 6 percent [air and a cement content of approximately 607 lb/yd3 . All test speci- ...-- 1.
wens were 7.9 by 3.9 by 2.6 in. Five specimens were prepared for every.
mixture tested.
34. The freeze-thaw tests followed the procedures of RILE COL. 2.
The test specimens were exposed to cycles of 16 to 17 hr of freezing at
-40 F and then I to 8 hr of thawing. Deiciug solutions of 3 percent
salt were maintained 0.08 to 0.18 in. deep on the block surfaces..
. ji Blocks wee .withdrawn ftrow testing when the.block surface had
19
:t .__..__A• • . _ _ __.._:. . . ... _--_._ _--_-_ __._._ •. . • . : '" ' . -" ' :: ' " " " ." . .-. . " , o ,, - : ," .'} ".", .; ," '
... . .1 . .. . . . . . . r,. , . '. :: ,- .. . ':,
0.70.6 5TH CYCLE £
0.5 _ LEGEND
00.4 - LB/YD3
0.3 10 U so CEMENT
0.2 °--__ 0S25TH CYCLE 9822
O.4 0 0 791L0.4 A a 659x0 A 463
0.2 3180.6 50TH CYCLE 30.4 "
0.3
0.20.01 0.1 1.0 10
WEIGHT LOSS. mgl/m 2
Figure 9. Relationship between water-cement ratio andweight loss (Clark 1980)
deteriorated to the point that the deicing solution could no longer be
* kept on the sample.
35. As shown in Figure 9, paving blocks with low water-cement
ratios suffered less damage in the freeze-thaw tests than those with
higher water-cement ratios. This figurte also suggests that blocks with
V high cement contents may have outperformed those with lower cement cou-. :- .
tents but the results-are not.conclusive. For a given, water-cemeut " -"
ratio the weight loss from fteezing and thawing did not change due to
aggregate type used in these tests.
36. No relation between absoq)tion and weight loss cau be identi-
fied ii Figure 10. Similarly, Clark (1980) found no relation between -
weight loss and either initial surface absorption. or density. In Fig-
ure 11 the effect of compressive strength on weight loss is less clear. "Samples that differed only -in strength due to curing for 35 days and
9 months showed no change in weight loss, uot did the Samples marked
'all other paving blocks" in Figure 11. However, two points i. iFig-
'i- ure it marked as low-strength paving blocks suggest a Strength relation-
ship. These two low-stregth. gesults also had relatively high. -
Hr 20
5. 6.28%0
wA 1 0cr A PAVING SLABHu49000 PAVING BLOCKS
cc4.7 0 045 25TH CYCLE 8 0
~4.300
4.1 0 00
00 03.7
0"1 0.1 1.0 10WEIGHT LOSS, nmiml .
Figur~e 10. Relationship betveen absorption andweight loes (Clark 1980)
L.EGEND
P -*VING $LAOqfA 0 PAVING OLCK AT,
~e~o 3~5 AYSw On
c a IbAT 9UONTI4all 0 A ALL OTHERI PAVING
ISLOCKS At 36 13AVS
. A PAVIUG DU=,K71 .01 101
isc 11. Relatinship N'twe'M ManT1 cimpr&efive atraItS~hFiur Aud vit.ige, (Clark 1980)f
I. 3&
water-cement ratios of 0.36 and 0.62. As shown in Figure 9, this has an
effect on weight loss. Since the increase in strength due only to curing
had no effect on weight loss and these two low-strength points had high
water-cement ratios, compressive strength does not appear to be a reli-
able indicator of weight loss for this test.
31. Current manufacturing practice produces paving block that are
subject to potential freezing and thawing damage because of inadequate
pore structure. However, freezing and thawing damage has not been re-
* ported as a major problem with paving blocks. The high cement content,
low water-cement ratios, and manufacturing process used to produce pay-
* ing blocks provide " final product that has high density and low perme-
ability. Even though the pore structure makes paving block potentially
vulnerable to freezing and thawing damage, this low permeability seems
to keep the paving block pore structure from becoming critically satu-* rated wnder the conditions that atost paving blocks encounter in the
* field. Resistance to freezing and thawing damage for paving blocks can-
not be set by strength or absorption limits. The block should be speci-
fied by requiring proven field performance or a laboratory freeze-thaw
test.
*38. An improved product 'could be produced by runufacturing paving&
Iblock with an adequate pore structure. In couventional concrete pave-,
* meits, this is provided by sound aggregates and entrained air in the
* cement paste. A more frost-resistant concrete paving block could-be de-veloped by providing a proper void atrueLure through air entraimat or
1 ~~possibly usiug hollow plastic mirshrs(0ylii nd-Sprinkel 1982.)or addiug crushed porous maeil Lt aad Skreda 1978).
skid resi5Lafice139. it getteral, skid resistance has not been a problem vith con
Ot~J crete block paveoents. Tite Nettierland speciiication NEN 1000 iticludes
a skid reslistance test. with the Leroux pendulum laboratory Apparatus
(Dreijer 1930.) li the United Kiagdoo the paviog block ieagrgt
Cannot contain tore than 25 ivrcent acid-soluble material (Dauson 1980).
This requirement avoids miaterials that polish easily under traffic Vitba resultiug decrease in paviag, block skid resist~ance. This bAs bcu
* . 1 *< f
found to be a particular problem with sea dredged material containing
shells.
Dimensional tolerances
-40. Paving blocks must be manufactured to close dimensional toler-
,.ances to simplify construction, provide tight joints, and ensure load-
. carrying and distributing properties of the completed pavements.
Allowable deviations in length and width measurements vary from 0.118 in.
n(3 m) in the German specification DIN 18501 to 0.059 in. (1.5 mm) in
the Netherlands specifications NEN 7000. The U. S. National Concrete
Masonry Association .(1979),allows a deviation of 0.062 in. (2 mm). Al-
I lowable height variations are less stringent and vary from the National
---- Concrete Masonry Association (1979) height variation of 0.125 in. (3 mm)
-'to the DIN 18501 height of 0.197 in. (5 mm).
I23
S.
1 K
4
-23
.. . . . .a
, . .•-,.,.. a '..'':'e. '-- -.k'- * ., ' ,-.,,' . , z.-..,o %,.
I • _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
.... . .. . .... . . ... ............ ............. ...-......-... ........ . .. . . . .. . . . . . . ..
Il
• • I-- --- - --- ------
PART III: BLOCK PAVEMENT CONSTRUCTION
41. Concrete block pavements require a subbase or base or both
just as conventional flexible pavements do. These layers are con-
structed and function the same for both flexible and block pavements.
42. A thin layer of sand is spread over the surface of the base
course. This sand acts as a laying or bedding course for the blocks.
In the United States, United Kingdom, and Australia the common practice
is to leave this sand uncompacted (National Concrete Masonry Association
1979, Lilley and Collins 1976, Morrish 1980), but in other areas the
sand layer may be compacted prior to placing the blocks (Kellersmann
1980, Working Committee on Concrete Block Paving 1965). Compacting the
sand before placing the blocks increases the effort of block laying
since the compacted sand layer will have to be screeded level again
after compaction and no sand will work into the joints from the sandlayer when the blocks are vibrated (Lilley 1980). The sand used in thelaying course should be a clean, well graded, sharp sand with a maximum
size of about 3/8 in. and a maximum of 3 percent silt or clay (National
Concrete Masonry Association 1979). Two recommended gradations for this
sand layer are shown in Figure 12. A cement mortar or a sand and dry
cement sand laying course can be used, but generally this is not recom-
mended unless the blocks are less than 2.4 in. thick (Meyer 1980) be-
cause the mortar would make it difficult to remove and reuse blocks and
because of potential frost damage to te mortar and the extra expense of
the mortar (Lilley 1980).43. Paving blocks are set by hand as shown in Figure 13. A multi- "
tude of laying patterns are possible; several were illustrated earlier
in Figure 3. Blocks are split to fit any cavities at edge of the pave- 11mnt, around manholes, etc., where a whole block will not fit A hydrau- . . -
lic block splitter is shown in Figure 14. Some work has been done to an- .
tomate placing paving block, but such systems are not in general upe now.44. The blocks are seated in the sund-laying course by vibrating
them with a vibratory plate compactor. Sand is swept from th surfa.o,
and as shown in Figures 15 and 16, vibrated into the joiuta.betwea the
ii :24
.7'
-%-
~~140I
CC
CAI /Iuj od0
~1* / o*:~ - j;';
kw'
CYIS~ .000~
25j
_ _ _ _ C
Figure 13. Placing concrete paving block (Z-blocks)
I Figure 14. Hydraulic block splitter.
K. - 426
INIR,1
Figure 15. Vibrating sand into paving block joints,
Berg Steel Pipe Plant.
27
blocks. After the joints are tightly filled, the excess sand is swept
from the surface.
45. Lilley and Collins (1976) suggest using a vibratory plate2with an area of 2.15 to 3.33 ft and a centrifugal force of 2,200 lb.
They state that a heavy compactor will not provide any additional bene-
fit. Kelleramann (1980) reports that a 3,300-lb vibratory roller or a
13,230-lb static roller are most commonly used in the Netherlands.
46. Lateral restraint must be provided on all sides of a block
pavement. There are no test data that provide actual limits on the
amount of restraint needed to keep blocks from separating under traffic.
47. Figure 17 shows the layout of a paving block work site. The
Figure 17. Construction at Berg Steel Pipe Plant
sand leveling course has been screeded out ahead of the working face.
Blocks have been~ delivered fromt the pallet to the working face where
they are placed by hand. Pallets of blocks are distributed downi the
face of the pavement advances. Lateral edge restraint is provided by
the oncrete strips wichw::: will lae eue f::az:uiZm:: :ou:::::::.48. lockpaveentcan be constructed by relatively unskilled
t 28
II
K _ _ _ _ _
~ j
I
labor. Production varies, depending on the complexity of the projectand skill of the block layers. Different estimates of output on block
pavement construction are shown in Table 3.
29
1:• .. i , i": i ' " i '/ : :- " " " ': ' "
...... ._ _ * 1.. . .......
PART IV: PERFORMANCE
Roads and Streets
49. Concrete block pavements provide an aesthetically pleasing
road surface that, if properly designed and constructed, is capable of
supporting heavy traffic. As pointed out in Part I, European nations
have had considerable satisfactory experience with this application of
paving blocks. At the U. S. Army Fulda Downs Barracks, Federal Republic
of Germany, 70 percent of the pavements have been-replaced with paving
block over the last 10 years. These pavements performed satisfactorily
under truck and M-60 tank traffic. Flexible pavement at one intersec-
tion required major patching every 3 to 4 months due to damage from M-60
tanks making 90-deg turns and was therefore replaced with paving block.
Inspection of this installation 9 months later found no evidence of
damage.*
50. Areas which lack continental Europe's history and experience
with small element modular paving have been more hesitant to use con-
crete block paving in roads and streets. Much of this hesitancy is due
to the lack of acceptable design procedures. The U. K. Cement and Con-
* crete Association, the Australian Cement and Concrete Association, U. S.
National Concrete Hasonry Association' and the U. S. Army Corps of Engi-
neers have issued preliminary or interim design criteria which should
* encourage further use of concrete paving block in roads and streets (Lil-
ley and Clark 1978, Lilley and Walker 1978, Hodgkinson and-Morrish 1980,
National Concrete Masonry Association 1980, Department of. the Amy 1979).
Rising prices for bituminous products,.'low maintenance cost of block'
pavements, and an increasing emphasis on a-esthetic and envirotmental
values may also help to increase use of block pavements in urban and.-
residential roads and streets.
51. Nuuicipalities in the United States have used paving block
* . *Personal communication, 1978, Mr. D. G. Frandsen, D1eprtatLo theAmy, European Division, Corps of Eagineers...-
30
i %, _ _.
V, '
for redevelopment work and also at intersections, bus loading areas, and
pedestrian crosswalks. The change in surface texture between conven-
tional and block pavements has been successful in discouraging drivers
from encroaching on pedestrian crosswalks. The block pavement roughness
limits vehicle speeds to a maximum of 35 to 40 mph and acts as an effec-
tive speed control. Some examples of these installations in the United
States can be found in Boston, Mass.; Brockton, Mass.; Providence, R. I.;
Baltimore, Md.; and El Cerrito, Calif., among others.
52. Approximately 300,000 ft2 of block pavement in Massachusetts i
and Rhode Island were examined as part of this project. These pavements
were up to 3 years old and were subject to the action of snow plows and
deicing salts in the winter and at two locations they were also exposedto salt mist from harbor areas. No environmental or traffic damage was
observed. There is one report of freezing and thawing damage to con-
crete block pavement in Colorado, but no information is available on the
quality of block. There is ample evidence in Europe and New England
that concrete paving block can be manufactured to withstind abrasion and
loads of traffic, deicing salts, and freezing and thawing damage.
53. Block pavements have also been used at railroad crossings.
The blocks within the tracks cannot lie immediately adjacent to the rail
without striking the railroad car wheel flanges. To solve this problem
at a railroad storage yard in England, angle-shaped bars were fixed on
the inside web of the rail (Miller-Cook 1980). The vertical leg pro-
vided edge restraint for the blocks, and the horizontal leg fixed to the
rail provided sufficient offset to allow passage of the railroad car
wheel flanges.
Industrial Applications
54. Paving blocks provide an excellent surfacing for a variety of
industrial applications and have been used by the Federal Republic of
Germany's industry for at least 30 years (Pesch 1980). They are used in
a variety of ground level interior or exterior production and storage
facilities in heavy industry, power plants, agricultural operations asi31*. .5,
' i ,i•',. - , ; " .... . i: •
. , .. ,. ,"• • .'. - • . " .- " ' " •. ','" .: : . - • :• T , ' ' '. i? •. \ ' , !: "
* ".. . " I' " .. "i " ' " "
.. . - - -7 - - - - '' -I T' 7-S I . - _
well as industrial roads, access roads, courtyards, open areas, ramps,
etc. Block pavement will not be suitable for applications which require
a clean dust-free floor such as assembly lines for electronic components.
Similarly, block pavements are unsuitable in applications such as dairy
facilities or meat storage where sanitation and hygiene requirements must
be met.
55. Industrial pavements can be subjected to very severe loads
from special material handling equipment such as forklifts and straddle
carriers, from highly concentrated loads in storage areas, or from im-
pact loading. The magnitude and intensity of this loading often exceed
any loadings which occur on nonindustrial roads or streets. Miller-Cook
(1980) describes several examples of the severity of these loadings on
industrial block pavements. These included a forklift carrying coiledsteel plate rolls weighing up to 23.1 tons and a measured load of
9.5 tons exerted on a 3-in.2 contact area steel wheel of a trailer. In
both of these cases 3.1-in.- (80-mm-) thick block pavements have per-
formed satisfactorily.
56. The Exposaic welded wire manufacturing plant in Mount Airy,
N. C., paved an approximate 20-ft width with Z-shaped concrete paving
block at the exit of a manufacturing line. Solid tire forklifts pick up
completed rolls of wire, exit from the plant onto the paving block, make
a 90-deg turn, and proceed to the storage yard or load the wire directlyonto tractor trailers. When inspected in 1979, the installation hadbeen in place for approximately 3 years and the manufacturer estimated
that a million tons of wire had been carried across the block pavement.
The blocks are 80-mm-thick Z-shaped blocks, lying on a 1-1/2-in. level-
ing course of stone dust with a 5-in. crushed stone base. The blocks
show no sign of settlement or chipping. There is some minor surface
wear similar to tl t observed on the portland cement concrete floor in-
aide the plant.
57. Industrial pavements can be subjected to very severe abrasion
:- - i -from solid rubber or steel wheel traffic, short radius turning or scrub .
bing by heavy vehicles, sliding of loads across the surface, etc, Spe-
cial paving blocks for very severe conditions have.been made with
tA32
OrV
selected abrasion-resistant aggregate or metal additives. These special
blocks have sandblast abrasion resistance as low as 3 cm3 per 50 cm2 of
surface area compared to the standard paving block's 15 cm3 per 50 cm2
(Pesch 1980).
58. Concrete block paving is highly resistant to fuel, hydraulic
fluid, oil, or similar materials that may be spilled in an industrial
environment. In the United Kingdom a cement-stabilized base is normally
used under block pavements in filling stations to reduce the possibility
that spilled fuel will seep into the subgrade. To further seal the sur-
face dry sand and cement are also occasionally swept into the block
joints. Debris will eventually lower the permeability of the joints,
but these precautions provide protection against subgrade and ground-
water contamination at the beginning of the pavement life.
59. Concrete paving blocks have also been used in pavements sub- I,jected to thermal stresses, and one German plant has stored red hot
steel coils directly on a block pavement (Pesch 1980). The blocks are a
special composition using blast furnace slag and slag cement.
60, Traffic on block pavements does not have to be delayed for
curing, and this offers some advantages when time for pavement repairs
is limited. Miller-Cook (1980) describes an application at a London
factory where an existing pavement could not be repaired without closing
the production line. The old pavement of granite, concrete, and asphalt
was removed, and a new block pavement was constructed in 3 days while
the plant was closed during a Christmas holiday. Blocks can also serveas a sacrificial layer where severe conditions will cause periodic re-
placement of the pavements. The modular nature of block pavements will
allow damaged portions of the pavement to be. easily removed and replaced
without grade changes.
61. Future plant expansion, need for future access to control and
utility lines, and anticipated settlement are additional situations
,, where industrial block pavements way prove useful. The Berg Steel PipeI Plant in Panama City, Fla., used between 35,000 and 45,000 ft2 of block
~4,pavement for a floor surface inside the plant. Cranes and manufacturing
machinery all have reinforced concrete foundations and the remainder .of
33 .
.. "
-" "~ .' o ' " ' .. .'... .. " ., . ' ' -': : .", --" . ,.,, . .,, .,.
the interior surfaces are 80-rn-thick Z blocks. The blocks are set on
a leveling course of sand over a thick layer of silty sand fill over a
soft organic subgrade. Future plant expansion is planned, and portions
of the block pavement will have to be taken up and relaid to allow for
this expansion and to allow access to buried control systems.
Port Facilities
62. Approximately 17 percent of the total cost of port container
*terminals goes for pavements (Van Leeuwen 1980). -Pavements in both gen-
eral cargo and container port areas are often subject to heavy concen-
trated loads; abrasive traffic; and spillage of various lubr-icants,
fuels, and hydraulic fluids, In addition port facilities are often
built on fill areas subject to large settlements, and block pavementsI
are economical uider these conditions.
'I63. The largest single application of block pavement has been at
Port Rashid in the United Arab Emirates where over 4 million ft2 of 100-
mm-thick block were placed in less-.than a year (Precast Concrete 1979).
The United Kingdom first used concrete block pavement for a port facil-
ity in 1975 at Cardiff for a 6450-ft2 loading bay, and as shown in~'ig-
ure 18 (constructed from data reported by Gerrard (1980)) -use of block
paving in U. K. ports grew rapidly in the neut 5 years. All of the
United Xiugdom pavement reported by Gerrar~d (1980) %Wed. 80-sw!-ttk
blocks.
J 64. The Europe Cootainer Terminus in R~otterdam has installed11.8 million ft of block pavement since 1967. The'pavewent consists of120-mm-ticik blocks, 2 in. of crushed rock or gravel, and 4.7 iu. of
cemtent-stabilized sand over sand fill (Van Leeuwe: 1980). The' *ntici-
pated settlement was 3,3 to 49ft during the frt10 years. Examples Iof some-of the loadings iziclude 11-ton axle loads which: are expected tosoon rise to 13.2 tons, straddle carriers with wheel- loads of 16.5 tons,
and 33- to 44-ton stacked containef loads transferred to the pavemoutthrough four corner supports'each with a 2.3-iti, contact area. Mander
these conditions block pavements have provided a.,highly satisfActory
- . I 34
v,~
OmA
200
180
NEW CONSTRUCTION *
160 CUMULATIVE THROUGHPREVIOUS YEAR
£140
0I
3 120
100
ZI~80w
wl
40i
~~01
1976 Iol 1917 1978 :1079 low0
YEAR U. . or 1i.i.F 1iguare is. list of paviuag block. aU .prs
35.
:,4
surface for 13 years in a port facility handling 850,000 containers a
year (Van Leeuwen 1980).
65. Pavements in port facilities face severe loading conditions,
* and block pavements have proven satisfactory for this application. Pat-
terson (1976) reviewed pavement requirements for container terminals and
recommaended surfacings shown in Table 4 for varying load and settlement
conditions. The surfacings considered were asphalt, tar, cast in situ
concrete, precast concrete slabs, concrete blocks, and staged construe-
* I hon. Local economic conditions may cause some variations in the order
in each category, and block pavements were generally most suitable when
settlement was anticipated.
* Drainage
66. The numerouf. sand-filled joints between individual blocks in
the pavement can allow water to soak into the underlying blase, subbase,Aand subgrade. This caa result in a reduction in strength of the sup-
j .. porting layers and unsatisfactory pavement performance. Weakening of aclay subgrade by infiltration of surface water is one suggested reasonfor the preuwture failure of a test item in an experimental installation
of a block pavement in the.U. K. (Barber and Knapan 1980), ond a sub-base Wakened by infiltrating moisture from a heavy rainstorm is blazud
for the 1976 failure of a block pavement in a part area (Gerrard 1.980)4,
67. It is widoly accepted but utiprovea that the block surfacebecowesi impenitable waider traffic as rubber, oil, .Aud other debris accu-
mulate in the joints,- Often a new block paVeOent is a$UatCJ to have
10 percent iflitial permability to surface water which is claimed to de-crease to near zero with time (Natioual Coucrete tiasonry Association
* 1 1979, Cruickshank 1976). In both of the proviously mentioned failures,the rains occurred while the block pavement vas new and the joints rela-
tiviely free draining. Wevertbhelesso it 4pp Ars to be unrealistically
optimistic to expect a block paveaent Lo become imprweable. For ia-
stance, Barber and Xuapton (3980) report from field observations in
* Europe that water poading oa the surface of &.block pavemnt will
36-
Wh.
percolate through the joint and into the pavement structure.
68. Surface infiltration is a recognized source of water in con-
ventional pavement structures. The Federal Highway Administration
(1980) recommends multiplying the design precipitation rate by coeffi-
cients of 0.50 to 0.67 for portland cement concrete pavements and coeffi-
cients of 0.33 to 0.50 for asphaltic concrete pavements to determine the
design water infiltration rate through these pavement surfaces. Gerrard
* ,(1980), based on his experience with the previously mentioned port fail-
ure, recommends that suggested infiltration rates of 10 percent for
block pavements be increased; and Lilley (1980), based on work by Clark
(1979), recommends not using moisture-sensitive material in the base of
the block pavement. Erosion and pumping of sand from the laying course
into shrinkage cracks of a cement-stabilized base are blamed for settle-
ments in an experimental block road in Denmark (Lesko 1980). Surface
infiltration is obviously a problem in block pavements and the assumption 4II that they become impermeable with traffic is not justified. There have
been suggestions to use bituminous seal coats or waterproof membranes on
base or subgrade layers and to seal the surface by sweeping dry cement
or dry clay into the joints, but no field applications of these tech-
niques are known.
69. Block pavements initially allow a high rate of water infil-
tration. This decreases with time under traffic and perhaps becomes
similar to a conventional pavement surface. Periodic measurements of
surface permeability of a test pavement in Australia are planned and
may provide more insight into this important problem (Sharp 1980). To!. reduce the effect of moisture, the Corps of Engineers limits the plas-
- ticity index of base and subbase material. in flexible pavements to 5 or I"' .. ~ less (Department of P ense 1978). Also all design strengths of the
subgrade, subbase, and base are selected from soaked California Dearing
Ratio (CBR) tests to allow for strength loss.due to moisture. Although
base and subbase material in pavements have generally been considered
free draining, current information suggests that these materials actu-
ally have low permeability and are saturated for long periods of time
(Nettles and Calhoun 1967, Cedergren 1974, Federal Highway
.',U .. V,47 -'"::
37
Administration 1980). Block pavements have essentially the same prob-
lems with moisture as conventional pavements and will require the same
surface and subsurface drainage considerations. Selection of material
properties to limit moisture effects and evaluation of soil strengths on
the basis of soaked test samples appear necessary.
Cost and Maintenance
70. Cost of paving blocks varies considerably due to local dif-
ferences in labor, material, block quality and size, and transportation
costs. According to interviews with several U. S. manufacturers in
1979, 80-mm- (3.1-in.-) thick block cost about $0.90 to $1.00 per ft2 at
the manufacturing plant and could be laid for $0.50 to $0.65 per ft2.
These rates are similar to 1978 rates of $0.86 per ft2 and $0.38 per ft2
quoted by Harris (1978) for 76-am- (3.0-in.-) thick blocks in Perth,
Australia. Transportation costs have to be added to these figures.
Table 5 compares the price for paving block in place from several dif-
, 4 ferent sources. The currency exchange rates in Table 6 were used for
all conversions. Direct comparison of these prices is also hindered by
inflation during this period and by other factors such as project size,transportation costs, and the extent of acceptance of block paving in
the local construction market.
71, When paving blocks are first introduced into a new market
area, initial construction costs are likely to be high due to contrac-
tors' unfamiliarity with paving blocks. In extreme cases contractors
may decline to bid on paving block, and in a few cases manufacturers
have had to lay their product themselves in a new market area. As the
product becomes familiac to construction contractors, bids becowe more
competitive. Table 7 is developed from data reported by Gerrard (1980)
and comparee bid -prices from six contractors for a large block paving
job at Dover, U, K. The coefficient o variation of the bids for the
block laying is half-that of the bids for construction of the conven- ,tional lean concrete base, indicatiog acceptance and competitiveness of
. the block pavements in that local construction market.
38
I4-
*! ". .I/ ;
II
72. Table 8 compares the reported initial construction costs for
several types of pavements. The block pavements are competitive in the
U. K. and Netherlands application, but are three to five times more ex-
pensive in the Australian application. This may be due to a variety of
factors such as different design loads, the changing relative price of
petroleum-based products from 1978 to 1980, the possible requirement to
design for settlement in port areas, or competitiveness of a well-
established paving block industry iD Europe. In the U. S., block pave-
ments are generally more expensive than conventional paving material,
but are often less expensive than decorative paving such as brick.
73. One advantage claimed for block pavements is reduced mainte-
nance. In 1969 the German firm Bauberaturg Zement Hannover conducted a
study of 87 block roads constructed between 1957 and 1969 to determine
maintenance costs and traffic damage to block pavements.* The traffic
damage results are shown in Table 9. Initial costs of block pavements
exceeded asphalt pavement by 4 to 14 percent, but after 10 years when
the asphalt was overlaid, the block pavement had lower total costs.
74. Other reports support these suggestions of low maintenance
costs. The block pavements at the heavily loaded and settlement-prone
container terminal at Rotterdam required maintenance one time in 10
years at a cost of $0.74 per ft2, which was twice the reported mainte-
nance interval and two-thirds the cost of maintenance for asphalt and
concrete (Van Leeuwen 1980). Experience in the Netherlands with block
paving in urban areas suggests that the blocks have a lifetime of
40 years and that once during this lifetime a full-scale repair will be.needed which will reuse 90 to 95 percent of the original paving blocks(Kellersmann 1980). Estimates of maintenance'requirements should beK
based on local experience, but there is often insufficient information
available in areas such as the U. S. where paving blocks are relatively
new.75. Sharp (1980) estimated costs in Australia for a 40-year life
* anufacturer's sales iterature describing this study translated andprovided by European Division, U. S. Amy Corps of Engineers. Now onfile at Office, Chief of Engineers (DAEN-PE-D), WathigLon, D. C.
39
A
"Y.'*
..... ..... ...... ''*_ _ _ _ _ _ _ _ _ _ _k'
U~ NQ.I
of 3sphalt, sprayed seal, and interlocking block pavements. A total of
36 pavements were ana].yzed with subgrade CBR's of 3, 8, and 20, and3 7
traffic levels ranging from 10 to 10 passes of an equivalent 18-kipsingle-axle load. These traffic levels are relatively low and are rep-
resentative of residential or minor urban streets.
76. The data from Sharp (1980) was used to prepare Figure 19,
2.00
~ 1.50BLOCK
00
z
!0.50-
SUBGRADE COR 8
PI6SES OF 18-KIP AXLE LOADS
Figure 19. Comparison of initial pavement costs
which compares the initial construction costs of the three types of
pavement for varying traffic levels. Only at the higher levels of traf-fic considered by Sharp does the block pavement become competitive with
W10 asphalt, and the sprayed seal always retains a cost advantage. In Fig-
ure 20, the initial construction and maintenance costs have been con-
verted to a single 40-year future worth, assuming an 8 percent interest
rate (Sharp 1980). No mnaintenance costs were included by Sharp (1980)for the block pavements, so a 20-year reconstruction of-the block pave-
ment was assumed, as suggested by Kellersmaua (1980). The cost of this
40
L2~
60
40
!I
BLOCK WITH MAINTENANCE
3DBLOCK
20
SEAL
10 1SUBGRADE CBR - 8
i.. niniiill I I m mI il I lilill I mitalli I aIIIIII
PASE OF 18,KIP AXLE LOAMSFigure 20. Comparison of 40-year future worth pavement costs
reconstruction was estimated assuming a 5 percent block replacement and
using the laying and block costs developed by Harris (1978) and de-
scribed earlier in this Part. This was added to the 40-year future worth
using Sharp's assumption of an 8 percent interest rate over the remain-
ing 20 years. When maintenance costs are included in the analysis, the
relative costs of the pavements change with different traffic levels
with each one becoming the most cost-effective at some point. Although
this analysis cannot be translated airectly to other areas, it illus-
trates the effect of several parameters in evaluating the cost of block
pavements. Maintenance costs must be included in all evaluations that
attempt to judge relative cost-effectiveness of different typos of
pavements.
Surface Properties
77. Block pavements are not as smooth as conventional pavements,)I41
I ,.- . W
and maximum vehicle speeds are limited to approximately 30 to 40 mph
(Lilley and Walker 1978). Table 10 compares minimum Army pavement
smoothness requirements with the maximum allowable surface deviation for
block pavements from two European sources. Block pavements meeting the
suggested European smoothness requirements will not meet Army pavement
requirements. For this reason the Corps of Engineers limits precast
concrete paving blocks to low-speed pavements, storage areas, and walk-
ways (Department of the Army 1979).
78. Skid resistance has not been a problem with block pavements.
The only known exception to this statement are two car parks in England
described by Lilley (1980). These car parks were constructed with
blocks that contained a large proportion of calcium carbonate in thefine aggregate, and these blocks polished under traffic within a few
weeks, providing a pavement surface with unsatisfactory skid resistance.U. K. Cement and Concrete Association and Interpave Specifications for
concrete block adopted a 25 percent limit of acid-soluble material in
the fine aggregate for paving block to avoid further problems of this
type. Kellersman (1980) reports that although many brick pavements havebecome polished and slippery under traffic, no concrete block pavement
has been replaced for skid resistance in Amsterdam during the 20 years
that concrete blocks have been in use.
79. Tests of skid resistance of concrete block have generally
found that they provide acceptable levels of skid resistance. Harris
(1978) reports that tests with the Transport Road Research Laboratory
portable skid resistance tester on paving block indicated very good low-
speed skid resistance. Tests with the California portable skid tester1on untrafficked paving blocks measured friction factors of 0.46 to 0.47compared to a minimum requirement of 0.30.* Figure 21, developed from
data reported by Meyer (1980) and Lesko (1980), shows that skid resis-tance of paving block descreases sharply under initial traffic, but this
effect lessens after the first year. Figure 22 is developed from data
"* * Personal communication, Hr. Dan Williams, Muller'Supply Company,Lodi, Calif.
42
1 •1
90
NEW LEGEND80 A VIBORG, DENMARK, BLOCKS
A VIBORG, DENMARK, ASPHALTIC CONCRETEAFTER 00 MELBOURNE, AUSTRALIA, BLOCKS
VIBRATION MEAN OF 5 TESTS
!2 \ kRANGE OF TEST RESULTS
S70
-Io" N
0
50
0 2 3
YEARS
Figure 21. Loss of skid resistance with pavement age
reported by Lesko (1980) that shows that the paving block skid resis-
tance decreases with age and increases in speed. Although the block
skid resistance is lower than for asphaltic concrete, it is above the
established Danish minimum requirement.
80. The skid resistance of a paving block surface is affected by
a large number of factors, iuch as tire characteristicsp vehicle speed,
surface moisture conditions, block surface texture, and quality of the
fine aggregate used in manufacturing the paving block. Experience to.g.; date indicates that paving block pavements
will provide an adequate ,.
skid-resistant surface for low-speed traffic provided that the blocks
lare manufactured with lish-resistant fine aggregate and have a tex-
t: tured surface.
43
.1
LI,
cc 0.7
* Fiure22. ~fetofLEGEND
OA~ ASPNIMTI CONCRETE ALU
0- Nil RANGE OF RESULTS.
/ _ _ _ _ _ _ _ _ _ _ __NEW
O'A'0 2YRS O1
PART V: BLOCK PAVEMENT TESTS
81. Continental Europe's long history of using brick, cobble-stone, etc., in modular paving construction allowed an easy adoption of
concrete paving block. The traditional design and construction proce-dures were easily modified for use with their new products, and there
was no need for testing to evaluate the performance of block pavements.
When the paving block industry spread to other areas without continental
Europe's background, questions arose concerning the design and perfor-
mance of block pavements. This led to a series of tests by different
organizations to evaluate the performance of block pavement and to col-
lect data on design for these pavements.
United Kingdom Tests
Laboratory plate load tests
82. The U. K. Cement and Concrete Association conducted plate
load tests of six different block shapes ranging from 2.4-in.- (60-mm-)
to 3.9-in.- (100-mm-) thick to determine the load-carrying characteris-tTd
tics of block pavements. These tests are described in detail by Knapton(1976) and Clark (1978). A 6- by 6-ft area was paved with paving blocks.
The blocks were on a 2-in.-thick sand leveling course, and lateral re-straint for the blocks was provided by timber curbs. Normal construction
methods were used to place-and vibrate the blocks. The bases used under
the sand course were concrete, crushed stone, and dense polystyrene.
- 83. Loads were applied to-.the block surface by a hydraulic Jack
and a 9.8-in.-diam Circular steel plate. An array of pressure cells lo-
cated flush with the base surface and directly under the sand layer
measured the distribution of-the load through the paving block and sand
layers. Figure 23 .hows' that the paving block does distribute loads,
and pressures measured on the base are significantly reduced. Figure 24
f ... shows that the shape of the block'had no influence on the load distribu-
tion of the block surface. Also, the thickness of the block had a rela-
tively modest effect va the load-distributing characteristics of the
Mlock pavements.
*. 45
-, , - t
171
LEGENDT
0 CONCRETE BASE, MEAN OF 8 TESTS,9 KNAPTON (1978)I RANGE OF RESULTS
00 CRUSHED STONE BASE, CLARK (1878)
70
~60
0 10 20 30 40 50 9D 10 so O0 10DAPPLED SURFACE LOAD. PWa
Figure'23.- Applied loads on the pavement base under block surfaces
4 *0 LOAD SPREADING FACTOR
STRESS ON BLOCK PAVING TO PRODUCE 1 mm DEFLECTION(02STRESS ON SUBBASE TO PRODUCE I mm DEFLECTION-
z0
0 a
50 10 70 890100
BLOCK THICKNE~SS, mmiFigure 24. Effects. of block thickness and shape.
(Clark 1978)
K L 46,
i ! I84. These tests demonstrated that the blocks develop side fric-
tion, rotational resistance, interlock, or some combination of these
factors to distribute vertical loads. Knapton (1976) used these results I
to develop an approximate design method by replacing the paving block
and sand layer with an equivalent 6.3-in. (160-mm) layer of flexible
pavement structure.
Field plate load tests
85. Knapton and Barber (1979) described a series of field plate
load tests on a paving block surface. The subgrade was an inorganic
clay of medium plasticity with a CBR of 2 percent. A granular limestone
base varied from 2 to 15.7 in. thick. Rectangular concrete blocks, 7.9
by 3.9 by 3.1 in., were laid in a herringbone pattern on 2 in. of loose
bedding sand. The blocks were vibrated into the bedding layer with 2 to
3 passes of a small vibratory plate compactor. Dry sand was then swept
over the block surface, and the blocks were revibrated. Precast con-
crete edge channels set in lean concrete provided edge restraint for thepavement test section. The final size of the block surfaced test sec-
tion was 10.0 by 19.7 ft.
86. The test section was repetitively loaded with 14,100 lb on a
7.9-in. square plate, and permanent surface deformations for vatying
numbers of load repetitions were recorded. Each loading was considered
to be equivalent to 30 applications of a standard 18-kip axle load. The
load was applied at seven locations where the granular base was 2.0, 5.9,
7.9, 9.8, 11.8, 13.8, and 15.7 in. thick.
87. When the load was applied to the section with a 2-in. base,
failure occurred immediately. Three blocks split in half, and the load-taig plate punched tihe surface outlined by the faces of the broken blocks
and block joints into the subgrade. Figure 25 shows the cumulative ver- .tical deformation measueed on the other base thicknestses at various uum-
- bars of load repetitions.
88. This test demonstrated that if an adequate base is used, a
block pavement could support heavy loads on a soft clay subgrade. Ob-
servations during this test revealed that initial settlements occur un-
til the blocks achieve iuterlock and thereafter further. deformation, y ,.
47
". . ..
_ _ _ _ _ I;
2.0 SIN
20
F0.8
4
5 10 50 100 Soo 1000NUMBER OF APPLIED.LWADS
Figure 25. Effect of base thickness and-load repetition
on vertical de~formation (Knapton and Barber 1979)
depends on the number of load repetitions, N ,described by the form,
y ac N .For Lhese tests ai averaged 0.37.Wa' ntaio et
89. Clark (1979) investigated the amouint of water that penetrated
a newly placed block surface. A block surfaced test'section was coni-
structed in a 6.6- by 6.6-ft concrete tank, as shown in Figure 26. Water
was applied to the block surface at different rates, and the amounts of
water that collected in the drainage channel and that penetrated theI
block surface and exited from the drainage outlet were mecasured. The
PAVING BLOCKSj
DRAINAGE OUTEr CON rTE TANK.'
Flare 26. Vater peoetratio tet bed- (Clack::1919)
48, 7
AMI
water was applied to the surface until the outflow from the drainage
outlet reached a constant rate for a period of 20 min. Each test gener-
ally took 45 to 60 min.
90. Table 11 summarizes the results of these tests. The mean
penetration of water through the block pavements with a slope of I per-
cent and clean sand in the joint was 19.6 percent with a coefficient of
variation of 14.7 percent. The limited range of block surface slopes
used in this test did not show any definite effect due to slope. Clark
(1979) postulates that the difference in water penetration between tests
I and 2 was due to accumulation of dust and fine particles in the joits
during the 28-day interval between tests. However, because of the inher-
ent variability of the test results, more data are needed to verify this
suggestion. An addition of clay dust to the joint filler sand in tests 8
and 9 reduced the water penetration to a wean of 11.5 percent. Clark
(1979) suggests that the sharp drop in water penetration in test 10 wasdueto he layparticles swelling during the 24-br delay between tests 9
and 10. Before Test 11 fine top soil was brushed onto the surface of
the joints rather than mixing with the sand joint filler to try to simu-
late the sealing of the joints with debris under traffic. This reducedthe quantity of inflow measured at the drainage outlet for about 50 mint,
but then 'flow increased until the penetration rate in Table 11 was
reached at a steady-flow rate.
91. This series of tests clearly shows that large water penetra-tion is possible in newly constructed block pavements. The tests do not
proid any infnto on pemeability of iu-1lc pavements which have
been subjected to traffic. This leaves the belief that block pavements
becomze iperiwable iIk~e to-accumulation of oil, rubber, aiid debris widertraffic without xperientl verification. ieisnlistl-
* Industrial road test
* tion of a paving block pavement at the exit of an industrial yard. ThL* test section was approximately 6.6 by 65.6 ft in area and situated so
that the wheels-on one side of each entering or exiting vehit'le had to
travel the length of. the block pavement *test section. This test section *
* 49.
4 07
was traversed by 57.3-kip four-axle articulated trucks, 41.9-kip three-
axle rigid trucks, and 33.1-kip two-axle trucks. The equivalent 18-kip
axle loads for each pass of these vehicles is 3, 1.4, and 1.6, respec-
tively (Barber and Knapton 1980).
93. The test section was constructed by breaking out a portion of
an existing concrete pavement. This existing concrete pavement provided
edge restraint on all sides for the block pavement test section. Ten
test items, each approximately 6.6 by 6.6 ft in area, were constructed
with varying base thicknesses and block shapes as shown in Table 12. A
.1 natural sandy clay was the subgrade for 6 items, a different sandy clay
was placed as backfill for the subgrade in 3 items, and heavy clay was
encountered in one item. Properties of these soils are shown in
Table 13. The water table was less than 11.8 in. from the subgrade
surface.
94. An unwashed and unprocessed local sand was used for the base
for all test items. The gradation of this material (Figure 27) shows it
is a poorly graded fine sand. Also shown in the figure is the gradation
usually required for a base material in the United Kingdom.
95. Two types of blocks were used to surface the test items,
Items 1-5 used a 3.9- by 7.9-in. rectangular block, and items 6-10 used
a 4.4- by 8.8-iu. shaped proprietary block. All blocks were 3.1" iu".
1'Wi . - .- " - - *--4: -. I . -
0 40
a :. .. .- . - - - . . . i _ _ 1 . I. , ..I ... _, . t .*- -li ..ll I I- ll-l
014rba• ""d ""-o7- .. -.- " - -I
. ., " "Figure 27. COdsrlra bs orega ation-
00" 6 •,1 " ' -1
• . -. 5:-L . -u<I&." . ':"i '."' -)"= =" = .... r .:i:,:-;ndu "• .r .' - l : .o.. b s co rs grada* .= io . , o- .... ,. - ,'l " - . ' .. • " : - :' ". , . . - . . " , .. . ,,: . . , : . .. . , , , , ' e',. a.n d , , .Xn a ,:,t.o,.'n- 3,,98 0 ),' : '
! ." . .., ... : . . . " .. .. ..-: : ' i -.: - ., > . . - ,.. . , . :. ,.. . : ." . ,: > ., . " :,: I- , , ,,.. " : • . ., , . . , , ' . .. . • • . , . , . ., . . , , , :, . -' . . • . . . . . , .. . . . . = , , , _ .= ,. ,
.1 _ _ _ _ . .: _ .... ..... ." .. •- -"',.. - . : _ __ _ _ _ _ ' ,." ! . .. .. . . . .. . " " .. . . . ": ' ' " - " " ," ' - " " " " - ' " - "" ' '
* thick and laid in a herringbone pattern. Construction of the block pave-
ment test section followed normal procedures. A 2- to 2.4-in.-thick un-
compacted sand leveling course was placed; the blocks were laid on this,
and then vibrated with additional sand vibrated into the joints from the
surface.
96. Deformations of the surface under traffic were measured peri-
odically to an accuracy of ±0.02 in. with an accurately leveled straight-
edge and a vernier caliper. Table 14 shows the maximum measured deforma-tion at various traffic levels. High initial settlements were recorded :in the first 300 axles of traffic with smaller deformations recorded
thereafter. Barber and Knapton suggest that the large deformation in
items 9 and 10 was due to inadequate compaction of the sandy clay back-
fill. and heavy rains which fell shortly after the test section was
opened to traffic.
97. Three different types of surface profiles, as shown in
Figure 28, developed under traffic. Item 1 had a general overall .
settlement, indicating densification of the base under traffic, but prob-
ably little or no shearing in the base or subgrade. Item 10 showed a
sharp unheaval of 1 in. above the untrafficked profile, indicating that
the base or subgrade was shearing. Item 8 had a similar profile with an
upheaval of almost 0.5 in. above the original surface. The remaining
items had profiles similar to item 5 in Figure 28 with two distinct ruts
but no upheaval above the original surface, It is uncertain whether
this shape is due to densification in two narrowly trafficked lanes,
shear movement, or a combination of these effects. .
98. Some of the conclusions reached by Barber and Knaptoo (1980)
from this test are:
.:. a. Inferior quality granular bases can be used with bloickt pavements on low-strength clay subgrades for light resi-
dential traffic.
b. The deformation of a block pavement is characterized byhigh initial settlements, followed by much smallerprogressive deformations which are approximately liaearon a logarithmic time scale.
£. Improved compaction could lower the high initial settle-meats in the substandard granular base.
51
Ak II
ITEM 1, 0 AXLES
ITEM , 0 XLESITEM T0, 0 AXLES
CONCR-CONCRETE: ~ ITE 1. M1300 AXLES
ITEM ±,30ALE . N
ITEM 5. 300 AXLES
LEG ENDI
-ITEM iI.... rM5
-ITEM 10
:1Figure 28. Surface profile from indust~rial road
d.A significant proportion of the initial settlement is dueto co~paction of the sand laying course.
e. The block pavemient is not watertight.. The base shouldretain its strengt4a whon wet, and the subgrade should beprotected by a waterproof membreane it it will losestrength at high m&oisture countents.
New Zealand Tests,,Cante rbury Circular Test Track
99. Seddon (1980) descr~ibes tha traffic tsnof two block test'sections at the Canterbury. .circular test. _track. -Tho'Cainterbury-virculartest track maoves an 8990-lb 1"ad ou twin truck tires around a circulartest- track that h"- a maean diameter of 30.2 ft. 'Taie~ PPlida
a velocity of 25.7 iopf ai.id is distributed over a 4.1-ft:-wide trafficV' lane. The natural subgradd at the test track is a greyvacke gravel de.-
posited by the Vaimakeviri Rivier. An'artificially weak test subgrade istot-w~ed by pl;aeing 0.47 in. of tive rubber between two filter fabrics ontop of the natural subgrade. This tire rubber consists of dusty flakes
.with a consisteacy similar to pipe tobacco. This artificial subgrade
52
gives high deflections with artificially steep deflection basins.
S100. During a test of nine different base course materials at the
test track, one base course failed after 150,000 wheel passes and had to
be replaced. The base course test sections were each 9.8 by 21 ft and
consisted of 5.9 in. of a granular subbase over the rubber subgrade,
followed by a filter fabric, 5.9 in. of base material, and a chip sealsurface. The failed base material was removed, and 2.4 in. of sand for
a laying course was placed on the subbase. Blocks 3.1 in. thick were
-.. laid in a herringbone pattern. Two different interlocking shaped blocks
were-u~ed and were identified as "Pavelock" and "Unipave."
101. One.hundred thousand passes were applied on the test track.
Figures 29 to 31 show the surface deformations recorded during these
tests. Resilient deflections decreased with increasing traffic. Rut
depth had high initial values and then increased more slowly with traf-
fic, and the transverse profile showed significant shear upheaval out-
side the traffic lane. The increase of density of the sand layer under
traffic is tabulated below;
! :Density, b/ft 3Center line Outer Area of Untrafficked
Block of Traffic Traffic Lane Area
Pavelock 116.6 115.6 106.9Unipave 114.6 117.4 104.7
0.I .
UN/PAVE 10 M. MEAN OF 8 BASECOURSES"
8i0 U) WHEEL PASSES x Ind
J4Figure 29. Beinkelanam deflections in conicreteocanunbound granular base course pavemeitts (Seddon 1980)
153
"V.4
--. ,, 0,09-
. " " . .. .. ... .. .. . .
I..2" ,, "jK': Ij" ",*U ml .O
-~ - ~ .-----.. ,'---.---- ---------- ~
0.75PA VELOCK
0 MEAN OF 8 BASECOURSES
8990 LB WHEEL PASSES x 1000Figure 30. Mean center line rut depths for concrete biock and
unbound granular base course pavements (Seddon 1980)
SAVERAGEERU D PTH BA5CUSEPA.MN
DOSANC FRMCFE IE N
Figr 31.5 Trnvrepoie-frtdpha 0,0wheLass(Sdo 90
percent wost 0e 2hi drope of to 20 3iiu 180 lbf 4 t
F ecetwiure 3,anserse trofi113.6 lb/t 3 weph pat d at0a0
13aportr coimpuctiontest. TNe Zelnd afndadntie 4402 the9 fabd-tinaosgetthat the sand was at abou 90enescen of 180l/twhncmate azer
tory density after the block pavement surface was seated with the vibra-
tory plate compactor.. Under traffic this deasity increased to over
k 54
*1~~~~... .____.___.___......______
II
98 percent of the laboratory density.
102. The Pavelock blocks had a compressive strength of 5,420 psi,
while the Unipave had a compressive strength of 8,210 psi. No blocks
showed any distress after 100,000 wheel passes.
103. An analysis with an elastic layer computer program was able
to match measured and computed deflection basins of the block pavement
by representing the blocks as an elastic layer of material with a modu-
lus of elasticity of 60,200 psi and a Poisson's ratio of 0.25. However,
the elastic layer analysis is not considered adequate by Seddon (1980),
and further analytical work is planned using a finite element program.
104. Some of Seddon's (1980) conclusions drawn from this study
are:
a. Concrete block pavements behave in a manner similar toflexible pavements with a thick surfacing.
b. On the basis of deflection basins concrete block pave-ments showed no greater stiffness than the chip sealsurfaced base course test items.
C. Excessive permanent rutting and possibly large initialtransient deflections were due to the sand layer.
d. The rounded beach sand used in the laying course was notsuitable for the loads used.
e, Improved methods of seating the block surface and com-pacting the loose sand laying course need to be studied.
Australian Tests
Road simulator tests
105. Shackel (1978) and Shackel and Arora (1978a) described a
series of tests on block pavements with the University of New South
Wales Road Simulator. The road simulator is a concrete trough which
allows construction of pavements 13.8 ft wide and 4.9 ft deep. A pave-
meat length of 14.8 ft is available-for testing. Simulated trafficI. loads are applied by hydraulic jacks to a series of seven 7.9- by 7.9-in.
steel plates to represent a load woving at 0.5 mph.
I .~ :- 106. Block and base thicknesses and loads examined in this test
are tabulated on the next page. Figure 32 shows a typical cross Section
55
", A l. ,7.
- i -____'____.........__ __ -_......... __ __,__ _____
Block Base Course ThicknessThickness 2.4 in. 3.9 in. 6.3 in.
3.9 in. X X X0 O 0
3.1 in. X X X0 O 0
2.4 in. X X X
NOTE: X -87 psi, O- 131 psi.
of the test section. A sandy loam subgrade was left in the concrete
trough from previous pavement tests, and additional sandy loam subgrade
material was obtained from the original source in Emu Plains, New South
Wales to backfill the existing test section to the new test levels. How-
ever, as shown in Figure 33, the new sandy loam backfill gradation dif-
fered from the original sandy loam subgrade slightly. Shackel and Arora
(1978b) reported that the original sandy loam was nonplastic, had a spe-
cific gravity of 2.64, had a modified American Association of State High-
way and Transportation Officials (AASHTO) optimum moisture content of
8.5 percent, and had a modified AASHTO optimum density of 125.9 lb/ft3.
The new sandy loam backfill was reported to have similar characteristics
but somewhat lower compaction test results (Shackel and Arora 1978a).
Shackel (1979) reports the subgrade CBR as 65 percent. Shackel and Arora
(1978a) reported that the dolerite base material did not differ signifi-
cantly from the dolerite base used previously by Shackel and Arora
(1978b). This material was described as nonplastic, with a specific
gravity of 2.87, modified AASHTO optimum moisture content of 6.5 percent,
and a modified AASHTO optimum density of 149.2 lb/ft3 . The gradation of
the dolerite base is shown in Figure 33. All blocks were identical
interlocking shapes laid in a herringbone pattern.
107. Data on permanent vertical deformation, horizontal movements,
resilient deflections, and pressure cell readings were collected after1,000 and 13,000 passes of the test loads. A typical transverse rut pro-
file is shown in Figure 34. Regression equations were developed for each
• I loading pressure to individually relate permanent vertical deformations,
7 :resilient deflections, and stress distribution to block and base
$: 56
1* I "2 -- " " . .- .+ .
, .,-.I. . . ., - . . .. - . :. . . . , . . . . . . ... ,",. : , , . - ,',.
3 EQUAL STRIPS 0 55.5 IN.
4. I.1.2 IN, THICK COMPACTED SAND 4.3_IN
8.7 IN. 2.4 IN. DOLERITE BASE
8.3 IN. 3.9 IN, DOLERITE BASEj
10.6 IN PRESSURE CELL 6.3 IN. DOLERITE BASE 10.6 IN.
BACKFILLED SANDY LOAM SLIBGRADE
17.7 IN.
EA R TH
PRESSURE
CELLS
ORIGINAL SANDYLOAM SAJBGRADE
Figure 32. Typical cross sectiou of w'ad simulator block test sectiou
E *14
- -~. -~ -- ---.- ----
-I j
zt
LLu
4~ 0
01 41
OaVIAAMM AAVI"
77
2i DISTANCE, IN.0 * 10 20 30 40 so5
~0.05-
I- 0.10
0'.15
a. 1,000 repetitionsj
WHEEL
0 10 20 30 40 50
0.10 -
THICKNESS,'-.IN.
0.20 LOCK OA.SE
0.26.5 0 2A421
$I PSI CQNTA CT PRESSURE
oaob. 13,004) ureoitious
Fi~uve 34. Typta wosv~ e cut profiles froa ~Oads*u-14tr tests (SIhackel-1978)
59
I? . ~ X
IS ''f
thickness; typical examnples* oftese for the 8- odae
log (rut depth) 3.867 - 0.988 log (block thickness)
-0.875 log (base thickness)
correlation coefficient (r) 0.82
or
log (stress at subgrade surface) =3.866 - 0.495 log (block thickness)
- 0.517 (base thickness)
r =0.93
108. After the tests to investigate relationships among block
thickness, base thickness, and loading pressure, another series of tests
evaluated the effect of block shapes. These tests found no significant
difference in performance between different interlocking shaped blocks,
but, as shown in Figure 35, rectangular blocks had somewhat greater rut
'1. depths than shaped blocks.
MOT6I BLOCK THICKNMSS 2A IN,
CO~NTACT PRESSURE. 01 M13AWQ WHEEL PASSES
0 RECTANGULARSC* a " {
0 10* .10 30 40 SO
F igure 35. Effect of block shape on defortutionsin the road siunlator (Shackel 1978)
*Theae equations are shiown~ in their original fore developed .us*ng S1units of uewasure. The tests used loading pressures of 600 and 900 kPa$block thickness of 60, 80, and 100 ow, and base thickness of 60, 100,
and 160 am. Units for theme equations are-aillisetres for depth and<.10 t thciuws an kilpascls fr stehsH __60
____ ___ '~--:- V
----- -__ -------. ~... .----.- - ---.--. -.---.-.--.---.
109. Another test examined the effect of reducing the loose lay-
ing course sand thickness from 2.0 to 1.2 in. As shown in Figure 36,
this led to a large reduction in permanent vertical deformations. There
was little effect on resilient deformation, but the measured pressure in
the subgrade increased from about 7 psi to 11 psi when the sand layer
thickness was reduced.
0.20 ~DISTANCE, FT 9 10 I
C - 0 5 3 6 0 1
LOOSE SAND
0.10 THICKNS
CTCKESSR*6 5
I N, A PI~IN USE THCKNESFigure 36. The effect of the sand layer thickness on
permattent vertical deformations (Shackel 1978)I
110. Some of the conclusions,(Shackel 1978) drawn from this4
study a. Concrete block pavements tead to perform as flexible 1.b. Block pavemeutts stiffen with increasing traffic and
achieve a "lshikedovn conditiou" after which further ac-cumulation of deformatiou is negligible.
c. Axk increaose in block or base tbickness improves
- perforsance.
d. Increases in block thickness are move significant, thanincreases in base thickUCessI
e. An increase in block thickness from 2.4 in. to 3.1 inimproves~ perfoxvu~wore than an increase from 3.1 in.to 3.9 in.
61
%V
. ,.77
111. These extensive tests and conclusions reported by Shackel
and Arora (1978a) and Shackel (1978) are strongly influenced by the
strong subgrade used. The dominant vertical deformation measured in the
tests must have come from densification rather than from shear movement.
This belief is supported by several factors. Shackel (1978) states that"except after the~ initial test (i.e., using those pavements in Fig-
ure 32) it was not possible to detect any significant rutting in the
dolerite base which could account for the development of the surface
rutting along the simulated wheel paths." An examination of the trans-
verse surface profiles in Figures 34 and 35 shows a general surface set-
tlement without any characteristic shear upheaval outside the loaded
areas. The reported mean dry density of the dolerite bases was 151.1 lb/3
ft (Shackel 1978) or 101 percent of the modified AASRfTO opcimum re-ported by Shackel and Arora (1978b). The original sandy loam subgrade
mean dry density was reported as 120.5 lb/ft 3(Shackel 1978) or 95.7 per-cent of the modified AASIITO optimum reported in Shackel and Arora
(1978b). The backfill saudy loam subgrade had a reported mean dry den-JI sity of 104.9 lb/ft (Shackel 1978). No compaction test data are re-
ported for this material except for the statement that the modifiedAASHTO density for the backfilled sandy loam was lower than the original
material but relative compaction levels were similar (Shackel 1978). TOprevent densification under traffic, current U. S. Army Corps of Engi-:
neers criteria (Department of the Army 1971) would require a minimum
(equivalent to modified AASHTO) and a 100 percent CE-55 density ill the
sification under traffic would occur ia the sandy loam.112., The small measured sraedeformations cosdeewith the
hihCRsubgrade, Shackel's reported observatious on rutting, the shape
of the transverse profiles, sandy loam compactiou levels,* and the reli-
Itively large deformations discovered -in the 2-in.-thick loose sand, lay-Iitig courses all strongly suggpst that derisificatioi and not shear was
the primary cause of surface defo mtions. Thorefore Shackel's (1978)regression equations and. conclusions caatbe appl ied 'to block pavewents
62 ... .... 1II
where shear deformations are occurring or compaction levels ace differ-
ent from those used in these tests.
Australian Road 2
Research Board Test Road
113. Sharp (1980) described a concrete block test.road being
built at the Melbourne headquarters of the Australian Road Research
Board (ARRB). Ten test sections 13.1 ft wide and 32.8 ft long are being
constructed between two parking lots used by the ARRB staff. The road I
will function as a minor residential road and is instrumented with pres-
sure cells and moisture gages. Extensive surveys are planned for road
user reactions (tire noise and roughness), surface deformations, skid
resistance, and surface permeability. The test sections are designed to I'Ievaluate effects of block thickness, base thickness, compaction levels,
and moisture content in the subgrade and base. The monitoring of the
test road began in 1981 (Australian Road Research Board 1981).
6oguth African T eatc, Heavy Vehicle Simulator -. o"
" 114. In 1979, the South African National Institute for Transport-.
and Road Research conducted a pilot 4tudy of the per-fowaace of inter-
locking, concrete block pavements under the- traffic of the-eav Vehicle-
Simulator.. Shatckel ('19M. .1980a) reportedt results,of this investiga-
- tion wIWI studied the effects o :.bokshape, block strength, laying
pattern, and block thickness.
1S, Pi, ure 37 shows the shapes of the block; to.te4l in this pro- ' It " " ., " . . , , . t ".. -- : : " ' .
gram. i he blocks were'laid on a 0.8-n.-thick layer of Saud iteer whic..
- lay-a 3,9-in, base course of poor quality ntuti gravel and a subgr4d -
W ith an. avea&ge C$& of- Z9 pervent. (Shackel 1980b). -"Prop eties of the -
'.4 1 . subgrade, base, and "and layer a.r ft a. ri d in Figure 38 an dTable 15.
iAovks were aI-Dd by haed, vibrated with a vibrakosy plate coiaacLor, Of-!
*.theu teVibratcd as sand was s ept into the joitits. The:quality Of blockft. laying in tems of jointwidth and wfotity as considered itnferior
4 . to Australian or Lureau .&'tai-drdo. but -represetatife...of South. Aic-.. ,
j layi standards ,hack..l 1979). Edg.e-rest.nt for, t.. ..... --WO 4-'-!- .
I . .. .. -, • i• £' . - . . - . N . N "A"" . . . ' , .
• " { .: ' .. . "" : . " , : . "\ . - " .' " " - ... " .' -N'n
. n _, _ _i_:..:,i( . : ,. " '" " , .. -- . .. ' - . - ' ":-.- J V-A- -:-,r5 .J. , J, ~ i: ( :
0'":N, "1- 7'. t -
. "w , ... . . .. . ": '. -. ( " ' " . . . '.",.' .-,..' ..' ' " ..'-- , .,. ' . ' " -' , 7
(NOT TO SCALE)
tE II Figure 37. Block shapes tested (Shackel 1979)
'I* to 3715U.S. SIEVE NUMBERlS WV0 1W0"'W" 40Y 20" ItV 4" '3~246 51*1 wJ ~ 7-J1111
~~1~ -A. 'U8G0J-.4 < 1tS
.~~~...~ tt'~ ~
404I~ ~~ ~ ~~~~~~~~~a Tspo~ddb noc~ oart ub.Ec ~~ eto a 92f
6.4 11
HIP.
2
contact area of 112.5 in. at a load of 8,990 lb. The wheel traversed
I -the test sections at a speed of 2.2 mph. During the trafficking of the 41
test sections vertical deformations of the surface and pressure cell
loadings were measured.
117. Figure 39 shows that block types C, D, and E of Figure 37
2.25
* BLOCK LAYINGSHAPE PATTERN
2.00-A[A STRETCHER B
. La HERRINGBONE1.75 B V
COz 1.50 STRETCHER A
9 BLOCK - 2.4 IN. THICKw1.25 EE 0 SAND - 0.8 IN. THICK
! - BASE - 3.9 IN. TH;CK.. 1.00 .
z
lz 0.76 '
WHEEL LOAD, KIPS .
8.99 11.24 ,13.49 15.74
0 10 20 30 40 60"t "
NO. OF WHEEL PASSES, x 103
Figure 39. Comparison of performance by block shape (Shackel 1980)
failed rapidly under the 8,990-lb wheel load, while type B was adequate
through the 11,240-lb load and type A was adequate through the 15,740-lb
load. Stretcher bond A in Figure 39 is the conventional stretcher bond
laying pattern in Figure 3a with the long side of the block perpendicu-
lar to traffic. Stretcher bond B is the same pattern but with long side
of the block parall:l to traffic to simulate conditions at an intersec-
tion where traffic would cross the pavement at right angles to stretcher
bond A.
* ' i65-
- 1-
• " "T' " .. .... . " "" : - .: : :' " ... : '::" :' ,'." " - - :- " " : .. ':;'-, ", . . .¢, :' ,.;; .;-',.r ,
.1 118. After the rapid failure of blocks C, D, and E new identical
test sections were constructed and trafficked with a load of 5,400 lb.
Block E again failed very rapidly, and the results of the 5,400 lb and
then progressively heavier traffic are shown in Figure 40. Table 16
summarizes the complete road simulator tests as reported by Shackel
(1979).
1.00WHEEL LOAD, KIPS
~ TIRE PR ESSUR E - 87 PSIBLOCK - 2.4 IN. THICK
0.75 - 8.99 SAND - U.8 IN. THICK3 0 11.24 BASE - 1.9 IN THICK
.1 1 a
2 0.25
0 10o0= 20.000 3.OOD 40,000NO, OF WHEEL PASSES
Figure 40. Effects of accelerated trafficking on 'block types C and D (Shackel 1980)
119. Shackel (1979) concluded from the results of these tests:
a. Block pavements tend to perform similarly to conveu-tional flexible pavements.
b. Most of the block pavements exhibited a progressivestiffening with traffic; i.e., the rate of deformationdecreased.
c. Pavements of block types A and 8 tended to achieve aninterlocked condition beyond which deformation accumula-tion was, very slow.
d. On1ce interlock had been achieved, increases in wheelload had little effect on deformation.~
a. The Shape of a paving block affects its ability toachieve interlock and support a given wheel load, aswell as affect.ing pavemen~t performance.
*.f. Shaped block perform better than rectangular blocks.
66.
g. Pressure cell measurements indicate that the beddingsand layer contributes to the structural capacity of thepavement.
h. The herringbone laying pattern tended to perform betterthan stretcher bond A or B.
i. Block strength within the range of 3700 to 7980 psi com-pressive strength and 610 to 1150 psi flexural strengthdid not affect the pavement performance.
120. The variation in subgrade CBR between values of 6 and
68 percent in Table 16 requires that conclusions c, d, e, f, and h be
considered with some caution until further experimental verification is
available. For instance, on the low CBR subgrades of 6 to 9 percent
all blocks (type E rectangular, type C interlocking on two sides, and
type B interlocking on four sides) failed under the 5,400-lb bond when
they were laid in the stretcher bond A. At the higher CBR values of 21
and 25 percent the block types C and D that interlock on two sides out-
performed the block type A that interlocks on four sides and that was
also on a higher CBR of 40 percent. All were laid in stretcher bond A.
Interlocking block types A and B were tested with either different lay-
ing patterns or significantly higher subgrade CBR values then blocks C,
D, and E under the 8,990-lb load. At similar CBR values (21-28 percent)
and with identical stretcher bond A laying patterns, the rectangular
block E outperformed block D (interlocking on two sides) and equalledblock C (interlocking on two sides). The importance of the subgrade
CBR is further borne out by Shackel's (1979) description of failure of
blocks C, D, and E under the 8,990-lb load: "Pavements quickly devel-
oped unacceptable degrees of rutting. As the rutting progressed, sub-
grade ±ailures eventually occurred and were manifested as heaving along
either side of the wheel path."
Danish Test Road
121. Lesko (1980) described a test road containing four sections
of concrete block pavement near Viborg, Denmark. Figure 41 shows the
four types of blocks included in the test road. In addition to the
iiA.-- 67 12 .':
-.. .-.b." . " .[
TestSetion C D EF
Name KSB-block FISK-block SF - block 1SF-block
Figure 41. Blocks and laying pattern used in Danishtest road (Lesko 1980)
block sections the test road includes two conventional asphaltic con-
crete sections and one section with an experimental asphaltic product.
122. Each block test item is 11.5 ft wide and 656 ft long. All
blocks are 3.1 in. thick. The pavement structure above the subgrade
consisted of a 15.7-in.-thicc sand subbase, 3.9-in. granular subbase,
7.9-in, cement-stabilized granular base, 1.2-in. sand laying course,
and the block surface. The cement-stabilized granular base contained
percent cement and had a 7-day compressive strength of 2,900 psi.I
Average block compressive strengths varied from 11,020 to 11,750 psi,
The saud layer was compacted by a roller; next blocks were placed by
hand. Sand vas swept into the joints between blocks, and then the blocks iwere compacted with a vibratory plate and a small tandem wheel rolter.
123. The test road was opened to traffic in October 1977. Lesko
(1980) reports that the block pavements have shown satisfactory durabil-
* ityo but there were some settlements over transverse shrinka-ge cracks
in the cement-stabilized material. This is believed to be due to ero-
There have also been come sporadic settlements close'to the edge of the
pavement that are believed to be due to laying techniques or lack of
124. The thickAess of the *aad layer decreased to 0.8 in. after
J8
2 years of traffic. There is also a slight tendency of joints in the
wheel pattern to widen slightly. The depth from the surface of the
block to sand in the joint is 0.4 in. in the wheel path compared to
0.2 in. outside the wheel path. Lesko (1980) concluded from the per-
formance of this road that:
a. Bitumen or cement-stabilized sand should be used forthe laying course.
b. In general the block pavements have performed remarkedlywell.
c. It is doubtful whether block pavements can be con-structed to a smoothness acceptable for main roads andmotorways, but block pavements can find extensive usein low-speed pavements.
69P
f.: j
iI
S j.. -.. ,
1. .
69
I
-,". -.. .- - -. ,- - ., ..
_______ ir-. .2:; --.ac. --
PART VI: WES TEST SECTION
Objective
125. In August 1979, the Waterways Experiment Station built and
tested a concrete block pavement test section. The objectives of the
test were to demonstrate the ability of block pavements to support
heavy truck traffic on a weak subgrade and to collect performance data
to help evaluate potential design methods for block pavements.
Test Section Description
126. The test section consisted of three test items each 15 ft
wide and 20 ft long. The native lean clay, classified as CL in the
Unified Soil Classification System, was excavated to a depth of approx-
imately 24 in. A plastic clay, known locally as buckshot clay and clas-
sified as CH, was placed and compacted to provide a low-strength test
subgrade with a CBR of 3 percent. The excavation was lined with a poly-
ethylene sheet prior to placing the buckshot clay to help prevent the
clay from drying out and increasing Ju strength during the test.
Items 2 and 3 had a 4-in.-thick subbase of gravelly sand and a 4-in.-
thick base course of crushed limestone above the buckshot clay. Item I
had a 4-in..-thick base course of crushed limestone placed directly on
the buckshot clay. All three itesu had a laying course of sand approxi- imately I in. thick, and the test itchns were surfaced with different
types of block. Lateral restraint for the blocks was provided by
10-ft-long, 4-in.-wide, 6-in.-deep oak ti'ers. These timbers were on
all four sides of each test item and each timber was anchored to the
grouad with three 24-iu.-long steel pins (Figure 42). Figure 43 shows
a cross section of the test section showing the three different test
I. I items.
127, The three types of blocks used in the4t tests are shown in
Figure 44. Item I used a E-slaaped block ("No. 1" in Figuret t4 hat was3.1 in. (80 wo) thick* ate r sd ectangular block ('111o. 2", inj
1z"
-i-ic
Figure 42. Timbers and pins used for edge restraint
Figure 44) that was also 3.1 in. (80 mm) thick; and item 3 used a shaped
block ("No. 3"1 in Figure 44) called Uni.stone that was 2.4 in. (60 mm)
thick. Table 17 shows the results of laboratory tests on the blocks.
* The blocks used in this test section were all of high quality and
strength.
128. Figure 45 shows the gradation, soil classification, and iI - Atterberg limits for the buckshot clay, gravelly sand, and crushed lime-
I stone. The buckshot clay is a dark brown backawamp, deposit of the His-
sissippi Riiver and has been used extensively in pavement investigations Iat the Waterways Experiment Station. Figure 46 shows the CoMVaCtioncurves for this material. Also plotted on these laboratory curves are
~field data from the test section. These data points are Identified as
data inside traffic lane." The construction data are qluality control 3I ~'tests run during construction of the test section. The other data werecollected from pits dug after traffic on the test section vas completed.
The inside traffic lane data were collected beneath. the traffic-loaded.
0*~"**I- 4~
El a~0
U4)0
0
~~ - U ~0tiU4)Ca
w ~:'. U)(LU
ci 4a:
v,,*, a:
It *1~~0 00
"4
.1-II~IJ~ . .1-i
I-. p..
11 8
I., <I:1
72.
j~ ~-~---~?a~ -
- -I
-A- ~ -- -.-. ~~-- ..--.-.-.-------------.-k
N*
7- 77"'?777*
Figure 44. Block shapes used i.n WES test
portiLon of the pavement; the outside traffic lane data were collected
und~er untrafficked portions of the test section. The field data densi-I
ties show some scatter, but there is no trend suggesting that compaction
of the clay occurred during traffic. Two field CBR tests in Figure 46show some drying with an accompanying increase in CR and are from the
buckshot clay surface in item 1. All other field CBR data points, in-'
cluding tests ran 6 in. below the buckshot clay surface in item 1, m~ain-~
aconsisteotly weak subgrade with a CDR of 3 to 4 percent throughout the
det.il effect of'surface drying in item .1 will be discussed in more.
129. The gravelly sand used in the subbase is a local alluvial 1material with smoth rounded particles. -Maimom density for this mate-
rial in f'igure 47 is at saturation, which is typical of cohesionless
soils. This gravelly sand twets the Corps of Eugineer requirements
(Veparttment of Defense 1978) for a flexibl e paveaeut subbase with a de-sign CM1 of 50 percenit.
130. The base eourse for all test items was crushed, vell-gradedI
limestone. The compaction4 curve kor thit material is shown in f7igure 48
aud shows au optimum CE-55 deusity of 147.6 lb/ft3 at an op t imWi
731
- . t
0 If
"1"1
~7
, 40
100___ vL0 E-55
90 CE-55 0 12 BLOW LABORATORY CE-12A 26 SLOW LABORATORY CE-26
8o Ci 65 BLOW LABORATORY CE-50 CONSTRUCTION FIELD DATAA FIELD DATA OUTSIDE TRAFFIC LANE
70 8 FIELD DATA IN TRAFFIC LANE
Ce60
tu 50- CE-26
-
0
40 I2011
0 MOI~iUI~ONENT.
110i
16 26rigue 4.. ~ ot.cla copaciou nd-CM u~e
f I
134 .~ :
132?
Ift]
( I0 2 4 a 10 12
MOISTUE ONTENT. PERCENT
Figure 47. CE-55 compaction curve for gravelly sand
UiS- -In
A Ee ~~~. ~ c oqa~o tuve for crushed lso
I ..-
moisture content of 4.3 percent. Laboratory CBR values for this material
consistently exceeded 100 percent. This material did not meet the Corps
of Engineers requirements (Department of Defense 1978) for a crushed
stone base course material with a design of 100 percent because the
plasticity index of 7 exceeded the mnaximium permissible value of 5 and
the 12 percent passing tbe No. 200 sieve (Figure 48) exceeded the maxi-
mum permissible value of 10 percent. However, these requirements are
primarily to limit the susceptibility of the base material to water.
Since the block pavement test section was inside of W'* Hangar No. 4 and
was not. subjected to raiu, these deviations in base quality are- not im- Vportant for this test.
Construction Procedures j131. The buckshot claiy was spread and leveled in 6-in, lifts with
a Caterpillar 1-4D-tracked dozer. Each lift was compacted with 8 cover-.J ages of a self-propelled Ingram pneumatic roller (Figure 49). The
I tt
ItI
IT
...t .fiur 4J'Rligo ev lysFrd,8cvF 0
.5po tt prsue,3 i
~~77
roller had seven smooth tires inflated at 65 psi and a gross load of
34,000 lb. The quality control data collected during construction found
dry densities to be slightly over 80 percent of CE-55 density and CBR
values consistently at 3 percent.
132. The gravelly sand and crushed limestone were spread and
leveled with a motor grader. Each layer was compacted with a self-
propelled Rex 700 vibratory roller (Figure 50). This machine has a
Figure 50, Rolling of liwegtoue base with-steel-Wheel vibrator~yrolle~r at approixiwately 14-kip dynamic forc
[ ;gross utti weight of aproxiMately 140000 lb and was operated at-the
aauf.cturer's dyuaic force ritg of 14 kips. Q.uality control data
collactcd'during contruct~ion foxuJA that gravelly esau4 dens1 ities Wert
lihtly above 95 pecut of.CE-55 dexiaity, aud the etushed limestage* vr. nlyabut 90 percentL. Rlk~ated Iase nit~ o rfied to im-wa olyabin passe of theilovrrsl e rolle fail th
p)rove these vdlues. These loo. densitir. are due to tht difficulty of
K bukshIot el'ay..
7'
-A
- - - .a.
133. The sand laying course was a masonry sand that was screeded
to a depth of approximately 1 ir., (Figure 51). Block laying followed
the procedures recommended by the National Concrete Masonry Association
(1979) and was described in Part III of this report. The construction
crew that built the test section had never worked with concrete paving
block before but encountered no problems during construction. Figure 52
shows the completed test section with item 1 in the foreground.
Traffic
134. Traffic loads were applied by the tandem-axle truck shown in
Figure 53. All traffic was low speed and estimated not to exceed 5 mph.
The truck was loaded with lead ingots until th. load on the tandem axle
was 25,000 lb. Wheel spacing for the tandem axle is shown in Figure 54..
Using the Corps of Engineers equivalency factors in Figure 55 (Brown and
Ahlvin 1961), one pass of the 25,000-lb dual-wheel, tandem-axle load is
equivalent to one pass of a standard 18,000-lb, dual-wheel, single-axle
load.135. The rope visible in Figures 52 and 53 marks the outside left
lane of traffic for the driver. No prescribed width or distribution of
traffic was used in applyipg traffic. The traffic distribution came
from the natural driving variation of the driver as he tried to drive
with the wheels adjacent to the marking rope. This led to a narrow
trafficked width of approximately 34 in. for each set of dual tires.
136. Traffic on a pavement tends to follow a normal distribution
shaped pattern within the trafficked area (Brown and Ahlvin 1961). Con-
sequently, not all portions of the pavement receive the same number of
passes or stress repetitions. Brown and Ahlvin (1961) developed a
method for calculating the relationship between the number of passes of
a vehicle and the coverages or maximum number of stress repetitions in
the traffic lane. This relationship assumes a normal distribution of
traffic and the other factors in the analysis are the wander of the ve-
hicle in the traffic lane and the spacing and configuration of wheels
and axles. For the truck used in this test the passes per coverage
79
14
k/
. . . . . . . . . . . ...... . . . . . ...
4~~~~~ Fiue5.Sceddsn
Fiigure 51. Scrcedd sand
WTI
Fiur 5. EStest setin items 1, 2, ad3 at 0 pase
80
.y1
VII.1A
Figure 53. Traf fic vehicle4
72 IN.
111 IN.4 N.
24 IN.
* Figure 54. Tire spacing for 25-kip tandem axle
81
.... ~ .....
41*
WJ ~ -
x , i >:.- I iw I-
W _j
C- 1crI
00 -',LLG
o 0
w 0
x-
IMO LO Lo Q
dl 'OVOl 31XV
82
',1
ratio is 1.05, or in the other words, the critical portion of the traf- 1
fic lane which receives the maximum number of stress repetitions is
loaded one time for every 1.05 passes of the test vehicle. All three
test items received 7500 passes of the test vehicle so the maximum num-
ber of coverages or stress repetitions at one point in the traffic lane
was 7143 (7500 + 1.05).
Performance
137. Figure 56 shows item I before traffic, and Figure 57 shows I,the item after 7500 passes (7143 coverages) of the test vehicle. The
item showed no sign of distress during traffic other than a few isolated
minor spalls (Figure 58) on the corners of a few blocks. "
Figure 56. Item 1 at 0 passes
.r
83
t t§
' 5
• , k
Fiur 57 Ite 1U1 at 7500 3U passes 4
W II
Figure 58. Item 1, corner break, 500 passes
84
-4 - - -----
_r T
_ ----- . .......
138. Figure 59 shows item 2 before traffic. Some minor rutting
is visible in Figu.- 60 after 1000 passes (952 coverages) and is more
pronounced in Figure 61 at the end of traffic after 7500 passes (7143
coverages). Minor corner spalls such as shown'in Figure 62 were more
common in item 2 than item I but. were still infrequent. Item 2 was
still serviceable at the conclusion of traffic.
139. Figure 63 shows item 3 prior to traffic. Rutting is visible
in Figure 64 after 500 passes (476 coverages) and is very pronounced in
Figure 65 after 4000 passes (3810 coverages). Tilting of the individual
blocks is visible in much of the traffic lane. The transition from the
landing mat surfaced vehicle maneuver area and the block surface at the .
north end of the west wheel path (rear left side of Figure 65) has
clearly failed and is undergoing major shear deformations. By 7500
passes (7143 coverages) in Figure 66 item 3 has deteriorated seriously.
A close-up view of the north end of the west wheel path iU Figure 67
,. shows the amount of surface deformation, block spalling, and block tilt- 4
ing that has occurred.
Figure 59. Item 2 at 0 passes
85
7?F'
U.........
Figure 60. Item 2 at 1000 passes
$4 4S4 - 4'
Figura 61. Items 2 at 7500 passes
86 ~
I P1
--- 4
Iva
Figure 62. Item 2, corner break, 3000 passes
n.i
* . . 0
Figure 63. Item 3 at 0 passies
87
J
KFigure 64. Item 3 at 500 passes
0I,
Figure 65. Item 3 at 4000 passes
88
77-'~7
1 7,1
1 111 1
Figure 66. Item 3 at 7500 passes
t- 1 4. 1~ le "I Z4 i~ A 4. . 4 3$ 46 -C44
I'LL
140. After traffic with the truck, the 11-48 tank in Figure 68 was
turned repeatedly on the rectangular blocks in item 2. This tank was
loaded with lead ingots for a total gross weight of 105,000 lb which is
representative of a current U. S. Army M1-60 tank. To make the turns,
one track of the tank was locked while the other track drove the vehicle
in circles. Af+tAr repeated turns, blocks showed no signs of damage orshifting, and 4t.he test was halted.
1Yieum 68.- 1-48 tank on item 2
14). Profiles of the block surface were re-corded ieosthe trans-
verse coater line of each item and along the longitudiual ceitter line of
the eastL amd Wes~t wheel paths. Viese profiles were recardgd before
traf fic, at intervals duriog traiffic., -and at the conclusion -of traf fic. jSimilar profiles were collected before zAnJ after triaffic on the surface
of the base, subbase, and suabgrde. These data-are sumnari2.ed in Pig-mires 69 to 77.
[ 142. Durijig the conattuction of the test section, &Iy densitytmoisture content, and CB1~values were recorded. At -the, conclusion of
90
II..j~
I SURFACE
B ASE
LEGEND
Da -a 70 stSS
SUBGRADE
I ~ ~ ~ ~ Fgure 69. Item I -crs etoipoie
'10\
*ore
II3fiur 70.1teI .40tdulPofls atw ealpt
92 0C ~L
SUBGRADE
3
U----0 0 PSAS
*- -00 noI PAssis
~Figure 71. Item.1 - 1oguioaljj profiltao west. wheel. path
SURFACE
BASE
1z .o--a03- 1 -a - - 1- - 01-"J
0 LEGENDD-0 0PASSES
O 7500?'Q PASSESSUDGRADE
Figure 72. Itecm 2IrNE setil proile
*194
BASE
OF1
SUB13ASE
aal
-a 011OI it
i oi-
j igiuIe.13. item 2 - onittiua piiils tost wheql path
BASE
' LEGEND I
nO - AM
F 1igure 74. Item 2 - oagituwiaij pojileo, vest w.heel. path I
R96
SURFACE
0
iI
04Y0 *~ ~ 10-hIfl
911
r'
4 4
- --
ND --G- -
4 SASE
I-4
S.9
~ ~F .~.. . .'---.-..--. --- -.--SURFACE ~ -
LEGEND0 0 PASSES
*7500 PASSES
5- BASE
4
3-
2-
2
.1 1 1 1 5 1 0 2
DWSA,4CE. VT
Figure 77. Item 3 - Ioagituwitaa profiles, vest~ w4iee path
99
a .. .. *. I
traffic, test pits were dug near the transverse center line, and similar
data were recorded inside and outside the trafficked area. A nuclear
density gage was used to record wet densities for the crushed stone and
gravelly sand, and dry densities were calculated using the moisture con- Vtent from oven-dried moisture samples. Density measurements for the
clay were made using a water balloon testing device. Tables 18 and 19
summarize the results of the density, moisture content, and CBR
measurements.
143. At the conclusion of traffic testing, plate load tests were
run on the block surface and then on the surface of the base course to.
determine a modulus of soil reaction (k). The results of these
tests are tabulated below.
k , pci
On Block On Base Block ValueItem Surface Surface Base Value
1 294 200 1.47
, 2 182 167 1.09
3 164 123 1.33
44Analysis of Result's
144. Item I gave the best performauce of all the items. Fig-
ures 69 to 71 show a small general surface subsidence of about 0.2 in..
in the trafficked area. There is no sign of any upheaval outside the
wheelpath. These profiles, along with the increase il crushed limestone
density in Table 18 from 89,4 percent of CE-55 outside tile traffic lane
to 100.3 percent of CE-55 inside the traffic lane, stroagly indicate that
the surface change was due to densificatioi in tile base.
145. Itev I with only a 4-in. base was designed as the weakest
test itew, but the surface of the subgrade dried out to form crust which,
increased the CDR from the original 3 percent to 6 or 7 percent. CHR
-measuretents taken 6 to, below the clay surface inside and outside tieI " :: ' traffic lpath gave resultt of 4 percent.. Hoisture co.tet in Table 18
I . also suggest that dryiug of the buckshot clay subgrade formed a Lhiu
. .. crust less than 6 in. thick in item 1.
* 1 1004i!
I * . " ' , - .; .. .... . . . . .,.'.. • , .- •. . . . . . . . ., -z,- . . - , . , . . . -,.:,*.., ' : e- w : "
146. The item 2 profiles in Figures 72 to 74 show a general sur-
face subsidence inside the trafficked area of 0.2 to 0.4 in. with ex-
treme values approaching 0.9 in. The cross sections in Figure 72 show
a small but distinct upheaval of 0.1 in. outside the traffic lane.
These cross sections also suggest that the gravelly sand was the sourceE i of the surface rutting.
147. The crushed limestone in item 2 (Table 18) shows no increase
in density between trafficked and untrafficked areas, although there is
a large increase in density between the construction and after traffic
data. This crushed limestone base never reached densities or CBR values
comparable to the base in item 1. The gravelly sand in item 2 showed no
change in density between trafficked and untrafficked areas but showed distinct reduction in density from the construction data. CDR values in
Table 18 are very low compared to the design value of 50 from laboratory
I "tests. This is partially due to the difficulty of running a field CBR
on a completely cohesionless soil where conditions of confinement and
surcharge are much different from the laboratory. Another problem is
the low density obtained iu the field, which has a direct bearing on the
148. Item 3 had the poorest performance of all the test items.
Rutting in Figure 66 is very pronounced when compared to the other items.
SThe crosu sectious in Figure 75 show a maximum surface 'ru 1.4 Xn. deep
and maximum upheaval of 1.3 in.. Both the base and subbase show large
defortations, but the subgrade shows relatively little deformation. The
. item 2 comeuts on the deu.sity and CDB values in Table 18 appear valid * -
for item 3 except the crushed limestone base traffic lane density andCDR values .showed a sharp increase over those taken outside the traffic
lane.149,. The Corps of Engineers 'designs flexible pavements by requir-.! I
" ing that each layer be protected against shear deformation by a speci-
icd thickness of material with a higher shear resistance as evaluated
by the CDR test, The design curve shomi in Figure 78 for 18,000-1b axle
Kloads is the one now used by the Corps of Engineers to select a thicknessof 'required superior material above a givea CER material for road and
i " .... " " " ' ........ ' . -T '--I ' - : /'" . "'"::: " :' - .. : ':: . ' ;% -' :! : 101. ,.
. ... •W!.. . .• - . "
4~ A- I
*1i
I_ - %
C.)
t. A
I.
street design. CBR design curves protect against a 1-in-, shear upheaval
above the original surface outside the- traffic lane. CBR desig, pro-
vides no protection against densification in the pavement layers. This
is handled separately by a compaction requirement that specifies tha
the field density must reach minimum percent of the optimum CE-55 Ths zra-
- 'tory density. The.required percent of CE-55 density varies with layer
depth, loads, and soil type and is available in the Army flexible pave-
ment design manual for roads and streets (Department.of the Army 1980).
150. Previous investigators have generally-agieed that block
pavements behave in manner similar to conventional flexible pavements,
and the profile rutting in items I to 3 siis tat surface rutting is
similar to rutting of conventional flexible pavements. :The Corps of
Engineers current design method for block pavements-(Department of the
Army 1979) follows the recommendations of Knaptoa (1976) and replaces
'" the sand and block layer with anequivale.c 6.5-in. thlckness. Then -
. the block pavement is .designed with the resulting equivalent paveent
thickness using exiuting flexible pavement CBR Curves.
151. Table 20 copares the-actual traffic n the test .sctio
with predicted capacity, using the C(R de-din with actual. thickness a.d
the equivalet thickness. The.predicted operations in table 20 were-
developed from Figure 78 which is based on vehi&- operations or passes .
being distributed in an l-ft-wide lane.. TV allow cou.0iSO on the.
saute basis the actual operatioos it failure in Table 20 aoe calculated
by multiplying the actuil coverage level by .. pass to couvage level of
1.13 for an 11-ft-wide lane as reconended by Brown and Ahlvin (1961) and
as used in Figure 78. The crushed liwestone bases were analyzed with a
CUR value of 100 pearent, consistent -with its laboratory test results
discussed earlier. The field test results in Table 18 show density and
3 CliR re3miuig the same or increasing from construction data outside of
traffic to the inside of-traffic lane. In item 1, the base did reach a-
. 100 percent CMi value. these data all confirm that the crushed lime-.
stone field performance agreed with the laboratory prediction. Aithoughthe gravelly sand subbase meets the laboratory te4uireoeuts for a 50 per-
cent CBR subbase, its field perfocmance proved tasatisfactoy. All
103
• ." , . ,'. ," .:. . ..,.',:,. '.
cohesionless materials depend on confining stresses to develop shear
strength, and this particular gravelly sand has given unsatisfactory sub-
base performance before when covered by relatively thin base and surfac-
ing (e.g.,. Ahlvin et al. 1971). The thin base, sand, and block layers
apparently did not provide sufficient confinement for the gravelly sand
subbase, The shear failure in the subbase is clearly visible in the
thinning of the subbase along the longitudinal profile in Figures 73,
74, 76, and 77, and the upheaval is visible in the cross section in
Figures 72 and 75. The decrease in density from the values of the con-
struction data in Table 18 to the value- for the outside of traffic and
in traffi: tests also suggests shearing occurred in the subbase mate-
rial. The subbase was given a design CBR of 20 percent for Table 20,
which is the lowest design CM value norm-lly allowed for subbase mate-
rials, Design CR values for the clay were selected from the in traffic
values in Table I.
152, Items I and 2 never approached the failure condition of
* - i-in. upheaval outside the traffie, lane, but item 3 reached this condi-
*.: ¢tio, after 5400 passes (5140 coverages or 5808 operations in Table 20).
An exawinAtion of Table 20 shows that predictions using the COR design,
_either @ith actual -hicknes2. or with equivalent thickness, did not antic-
ipate actual vtdoinrance .curately. Without the use of the equivalent
thckAness app oiwatiou the COB design method is very couservative. For
_. all cases Of predicted performonce the clay layer is critical, but under
acetual traffic.Ue clay never showed iny sign of signtificaut shearing.
- ilure- of itta 3 was due to the failure in shear of the subbase, which
is not adequately represented by the reduction of the CUR to 20 percent.
".Au appreciably lower number is necessary, but the efective CUR cannoth be determined fcra the traffic tests because the actual effect of the
block surface is unkutan.
153. Under similar traffic conditions the 3.t-io.-thick block id.
item 2 did much better than the 2.4-io.-thick block iu ites 3. The
shaiped blocks in items I and 3 gave much bigher ratios of the k o"
.ttf block surface to the k on the base course (tabulated in paa-
. . graph 143) than on the rectangular bl4ak in item 2.. This suggests som
104.. ,. .
.........-..........
advantage in load distribution characteristics for shaped blocks when
compared to rectangular block. However, under traffic the advantage of
the thicker block in item 2 outweighed any advantage of the block shape .1
in item 3.
Conclusions
154. The following conclusions can be made:
a. Block pavements behave in a manner similar to flexibleI1 pavements.
b. Block pavements can be designed and constructed to carryheavy truck traffic over a soft clay subgrade.
c. The 6.5-in. equivalency adopted by the Corps of Engi-neers (Department of the Army 1979) for block pavementdesign will allow a conservative design but will notpredict actual performance.
d. Limited data suggest that increased block thickness can
outweigh improved load distributing characteristics ofshaped blocks.
-K- -
105
4 3
-----------------------------------------------------
-- ...- -.......----
PART VII: ANALYSIS AND DESIGN
Behavior
155. A concrete block pavement functions as a flexible pavement.
Loads applied to the high-quality block surface are distributed through
progressively lower quality materials to the subgrade. Both interlock-
ing and rectangular blocks effectively distribute load to lower layers,
as shown in Figure 23. The plate load tests tabulated in paragraph 143
also indicate that the block surface has load-distributing capability
and extensive pressure cell measurements under shaped blocks reported by
Shickel (1978) provide further confirmation. The block test section
rutting described in Parts V and VI of this report are consistent with
flexible pavement behavior.
156. Although the block pavement behaves as a flexible pavement
in general terms, there are significant differences between an asphaltic
concrete surfaced flexible pavement and a concrete block surfaced flex-
ible pavement, In a multilayered elastic system the load-distributing
characteristics of any layer increases as the elastic modulus of the
layer increases. However, in the case of a thin surface layer with a
much higher modulus than the next lower layer (thin concrete over a 'T
thick granular base, for instance) the maximum horizontal shear stresses
at the middle of the stiff layer and tensile stresses at the bottom of
the layer increase significantly (Yoder and Witczak 1975). This distri-
bution of stresses is quite different from a conventional asphaltic con-
crete flexible pavement.
157. The block layer is not an elastic layer, but consists of mod-
ular units separated by sand-filled joints. Consequently, the load-
distributing characteristics and stress levels in the blocks themselves
cannot be calculated with elastic theory. Pressure cell measurements by
Shackel (1978) showed some indications that pressure measurements under
the block surface decreased with load applications. This may indicate
* that an increase in the stiffness or load-distributing ability of the
block layer occurs as the blocks wedge more tightly together or inter-
lock under traffic.
t:. 106
158. Block pavements distribute loads in a manner similar to con-
ventional flexible pavements, but their modular nature makes it exceed-
ingly difficult to evaluate them analytically. Several investigators
have attempted to reduce the block surface to an equivalent thickness of
conventional flexible pavement material for design. Although block pave-
mentp, are a flexible pavement, their modular nature, individually high
block elastic modulus values, and possible increasing stiffness under
traffic make them significantly different from conventional asphaltic
concrete flexible pavements.
Block Shape and Laying Pattern
159. The plate load tests conducted by Knapton (1976) and Clark
(1978) found no difference in the load-distributing ability of rectangu-
lar and shaped blocks. These test results are shown in Figure 24, and
Lilley (1980) states that block pavement performance in the United King-
domsine hei tetsconfirm these results. Shackel (1979s 1980) con-
cluded from his trafficking tests in Australia and South Africa -that
shaped blocks that interlock perform better than rectangular blocks.Hlowever, the large range of subgrade CBR values in Table 16 requires ad-ditioual experimental data to confirm this conclusion.
160. In the WES trafficking test (Part VI) rectangular blocksPerformed well, and in the Netherlands rectangular blocks have a longbistory of acceptable use (Kellersman 1980-t Van der Vlist 1980). In
Table 14, items 4 and 7 and items 5 and 6 differed only in block shape,
but there was little difference in their deformation wider traffic. The
information that is 3V4ilable ROW indiCates that either rectangular orinterlocking shaped blocks can be used in pave~ents, and no differencein performance has been proven.
161. The most cowon -laying patterns for block pavements subjectto vehicular traffic are either herringbone or stretcher bond (see Fig-
ure 3). Shackel (1980) found in trafficking tests that there was someindication that herringbonet laying patterns outperfomd stretcher bond,but he felt more experimental datee needed to confirm this. Lilley
107
4
S- ..... ....-
and Collins (1976), Lilley and Clark (1978), and Lilley and Walker (1978)
only recommended using the stretcher bond and herringbone laying pat-
terns under vehicular traffic. Also, based on observations of block
pavements in the Netherlands and in a car park in the United Kingdom,
Lilley (1980) recommended that rectangular blocks subject to vehicular
traffic should always be laid in a herringbone pattern to avoid creep.
This creep displaces the rectangular blocks laid in stretcher bond in
the direction of traffic. With the information now available it appears
that rectangular blocks subject to traffic should be laid in a herring-
bone pattern, while interlocking shaped blocks may be placed in either
herringbone or stretcher bond patterns.
Block Thickness
162. The plate load test results in Figure 24 from Clark (1978)
show that as the paving block thickness increases the load spreading
ability of the paving block surface increases. However, the effect in
this figure is modest and there is a significant amount of scatter in
the data. Knapton (1976) did not feel that the effect of block thick-
ness could be quantified and recommended use of the 3.1-in.- (80-mm-)
thick block for all but the most lightly trafficked p.vement. This rec-
ommendation was based on the experience and recommendation of continen-
tal Europe.
163. On the basis of traffic tests on the high-strength Austra-
lian test subgrade, Shackel (1980b) concluded that block thickness was
a major design factor. Block thickness affected rutting, surface de-
flections, and subgrade stresses. The W traffic test also showed
markedly improved performance of the'3.1-in.-Lhick block-in item 2 over
the 2.4-in.-thick block in item 3 with similar support conditions.
4 164. Block thickness his been ashown to have a decided effuct on
block pavement performance in'the United Kingdom plate load tests, the
Australian road simulator teats, and the VES trafficking tests. Ho-
ever, the only method devgloped to aualytically evaluate the effect of
block thickness is Shacke's (1978).vegression equationa, which as
:4' v1. -
. . : ....' : !.
=. -
_ ~ v . ,. o.. ,m ,,. , -..., -,, .....-__"
discussed earlier cannot be safely extrapolated to conditions different
from the Australian road simulator test pavement.
Laying Course
165. The sand laying course of I to 2 in. used underneath the
paving blocks is a construction necessity to allow the final leveling of
the block surface, but it is an element of structural weakness in the
pavement. Although Shackel (1979) found that a sand laying course con-
tributed to the reduction of stress in the underlying layers, Figure 36
(Shackel 1978) and the tabulation in paragraph 101 (Seddon 1980) clearly
show that the sand layer as presently constructed densifies under traf-
fic. This has to result in a deterioration of the smoothness of the
pavement surface.
166. To avoid or at least minimize this, the sand layer thickness
4 . should be kept to a minimum. The recommendations of Shackel (1978) andtheir successful application in the WES traffic tests indicate that the
thickness of the sand laying course should be limited to I in. If this
sand layer is properly compacted, an improved surface must result.
Either the sand layer can be compacted better prior to placement of the
blocks with acceptance of the problems noted by Lilley (1980), summa-,rized in paragraph 42, or else the paving blocks can be compacted withheavier and more effective vibratory compaction equipment to limit densi-
fication of the sand under traffic. Tests must be conducted to deter- :
mine the effectiveness of these suggestions.
Design 1 >167. Initial design methods for block pavements were developed
from continental European experience. These designs were essentially
qualitative and specified base and block thicknesses that had previously
proved adequate for similar subgrades, traffic, and climates. When Iblock pavements began to spread to areas without continental Europe's
experience with block pavements, :users began to demand design methods
109
..................................... ........
that could be used with confidence in these new areas. Also, as blocks
began to be used for road and storage areas subject to heavy traffic
loads over poor subgrades, the need for improved paving block design
methods became apparent. A design method for block pavements should
provide a safe, economical pavement and must account for variations in
subgrade strength and traffic loadings.
168. Block pavements generally fail by rutting, and individual
block breakage has been a minor, isolated distress. The high-quality
blocks now in use have proven adequate to withstand the stresses gener-
ated by a variety of heavy loads. The block surface distributes the
applied load over a larger area which reduces the stresses on the lower
pavement layers. Rutting in block pavement is caused by densification
or shear deformation in the underlying base, subbase, and subgrade. The
allowable rutting in a pavement can vary, depending on the use of the
surface and desires of the user.
169. Densification can be controlled by requiring adequate com-
paction in the pavement layers and existing flexible pavement criteria
found in current design manuals (Department of Defense 1978, Department
of the Amy 1980) should be adequate. Protection against shear is a
more complex problem since block surfaces distribute load over a larger
area than asphaltic concrete surface. Consequentlyt existing flexible
pavement design based on experience with asphaltic concrete will be cons-
ervative. Figure 79 compares failure data extracted from Figure 25 and
-Table 16. From Figure 25, failure was taken as 1-in. vertical deforma-
tion, and in Table 16 only data from the first trafficking of a section
was considered. Figure 79 also contains data from the WES tests in Fig-
ure 23 and the Australian tests sumarized in the tabulation in para-graph 106, and these data were all plotted as not failing since rutting
was either much less than 1 in. or occurred due to a weakness in a pave-I ment layer above the subgrade. The data from Seddon (1980) were not
used because the strength of the subgrade was unknown. The data from
Barber and Xnapton (1980) were not used since deformation contributions
from the subgrade and poor quality base could not be separated.r 170. From Figure 79, it appears that a separating line between
110
- - o -0!. . ~.. , .~
.].- -
*o 1.. 0.-
U) w
I w
lBl
.... ww m~ _31 -14
V, C' ew 2W . 00oz Ww~ I r-
u~ u 0 IL 44j2i~i004
-) o O m I- w ~ww 0 .56 i 0v
w *s Uu
mfaaIN I a
*NI ~S3N~SH a
failure and nonfailure for 500 passes could be estimated, but there is a
significant gap in the data. The data also show that rectangular block
test results are clustered to the left of the diagram, indicating severe
test conditions of low CBR or high wheel loads, while shaped block test
results are clustered to the right of the diagram, indicating less severe
test conditions. This further illustrates why claims that shaped blocks
outperform rectangular blocks cannot be accepted from the test data pres-
ently available. The minimum thickness for a block pavement will be
about 7.4 in.: 4-in. base, 1-in. sand leveling course, and 2.4-in.
block. More test results are needed to fill in the gaps in the existing
test section data before more extensive analysis can be expected to pro-
vide a separate empirical design for block pavements or an empirical
adjustment to existing flexible pavement design.
171. Selection of a design method for block pavements is mainly
limited to methods based on previous experience, Shackel's (1980b) pro-
posed method based on regression equations (Shackel 1978) developed from
field tests, the National Concrete Masonry Association (1980) method
that uses conventional CBR flexible pavement relationships, or the modi-
flied CBR flexible pavement design method that allows increased effective
or equivalent thickness for the paving block and sand laying course
(Knapton 1976, Department of the Army 1979). The design methods that
rely on previous experience with similar subgrades, pavement.materials,
traffic, and climate cannot be used unless all of these conditions are
duplicated at the new site. The experience-based methods also tend to
be conservative and are probably most appropriate for lightly loaded
pavements. As
* .discussed in Part V, Shackel's regression equations cannot be applied
to protect against shear deformation or for conditions different from
the Australian test track conditions, Consequently, their use cannot be
justified for design. The direct adoption of conventional CBR flexible
* . pavement design curves fails to give any weight to the increased load-distributing ability of the paving blocks. The use of. an equivalent in-creased thickness with CBR design curves as originally proposed by Knap-
ton (1976) and later adopted by the Corps of Engineers (.epartvent of
ii i, . tS 112
14 .5'T
- -g1
the Army 1979) offers a useful design expedient that recognizes the
increased load-distributing ability of the paving blocks. However,
the existing equivalent thickness that is used is based on simplifying
assumptions and limited testing that could be improved. With the infor-
mation currently available, the existing modified CBR flexible pavement
design methods using an equivalent thickness appear to be the most prac-
tical. The recommended design method consists of first developing a con-
ventional flexible pavement design using the Corps of Engineers CBR
method and then treating the sand and concrete paving block as equiva-
lent to 6-1/2 in. of asphalt concrete surfacing and base material. This
will provide a conservative design but cannot be expected to predict
actual performance.
172. Shackel (1980b) has proposed a design method for pavingblock on stabilized material by using elastic layer analysis. There are
no data presently available that will allow evaluation of these condi-
tions. Until more data are developed, the established Corps of Engi-
neers equivalency factor for stabilized materials (Department of De-
fense 1978, Department of the Army 1980) can be used to conservatively
design paving block pavements with stabilized layers. Design for frost
conditions should use the current Corps of Engineers flexible pavement
methods (Department of the Amy 1965), but actual thicknessesinstead of
equivalent thicknesses should be used for calculating frost protection.
Block Strength
173. Block breakage has not been a major cause of block pavement
-distress even though Seddon (1980) has trafficked blocks with compres
sive strengths as low as 5420 psi, and Shackel (1979) has used blocks as
low as 3700 psi. This suggests that the block strength requirements in
Table 2 are conservative for traffic loads, but these high strengths may
be needed for handling and abrasion resistance. If lower strengths areIachieved through reduced cement contents, the consequent increase in
the water-cement ratio will tend to produce a concrete of greater per-
eability which will allow critical saturation to develop faster when
113
I!
- ~~ ~~~ ...... . .- -- . . ....-.
the product is exposed to water. This concrete will be more susceptible
to freezing and thawing damage unless it is protected by an adequate
pore structure such as provided by proper air entrainment. These fac-
tors must be studied in more detail before lower strengths for these
products can be accepted.
1141
PART 'VIII: CONCLUSIONS A1ND RECOMMENDATIONS
Conclusions
174. The following conclusions are made:
a. Concrete paving blocks are a proven pavement materialcapable of providing an aesthetic surface that can sup-port heavy loads over soft subgrades, requires compara-tively little maintenance, and allows easy access to
J utilities or similar items below the pavement surface.*Concrete paving block can only be used for low-speedtraffic. They have a high initial cost and low mainte-nance cost.
b. Current Corps of Engineers method of design (Department-of the Army 1979) using an equivalent thickness appears-to be the most practical method of achieving a conserva-tive design but will not accurately predict actual blockpavement performance.
c.Either rectangular or shaped interlocking blocks can be
used for pavements, but rectangular blocks subject toI~1 -. vehicular traffic should only be laid in a herringbone
pattern while shaped interlocking blocks can be placed
in either a herringbone or stretcher bond pattern.di. Specifications for paving blocks should include the
following:*. (1) Btegh locks with a compressive strength of
_Approximately 800pihave a proven history o
(2 R~itance to trezing~~wn Either a
proven histr of perf rance under similar siteIcouditions or a laboratory freeze-thaw test shouldbe required.
_V_
"( feliOmensional tolerance. Standard of the local in-us~ty such as National Concrete tiasonry Associa-
~. tion (1979) in the U. S. or DIN 18501 in, Germanyshould be used. A o1itetrr
(4) rees4 tace. Amdfe sandblastteto0 tti~r abrasio i test methiod should be developed to
Wtest :for thi , decttve surf aqes on the block.
Recoimmedations175. The folIkowiag recomendatioas are- made:
a.Additioual trafficking tst and aalysi6 feLsi~LI; 115
p q 7 _7j4 :
pavements are needed to develop a more accurate andless conservative'design procedure. Development of animproved Procedure by the Corps of Engineers does not.appear warranted unless use of block pavements by theCorps increases considerably.b. The effects Of using lower strength blocks is worth
further study and field trials,S. -Construction procedures to automate block placem~ent andto improve compaction of the sand bedding layer shouldbe investigated.d. Methods of reducing initial block pavement perxmeability
should be developed.
1161
REFERENCES
Ahlvin, R. G. et al. 1971. "Multiple-Wheel Heavy Gear Load PavementTests," Technical Report S-71-17, 5 Vol, U, S. Army Engineer WaterwaysExperiment Station, Vicksburg, Hiss.
American Concrete Institute. 1969. "Recommended Practice for Concrete
Floor and Slab Construction," ACI 302-69, Detroit, Mich.
Asphalt Institute. 1979. "Chicago's State Street Is Now an Asphalt-Paved Transit Mall," Asphalt News, Vol 2, No. 4, College Park, Md.
Australian Road Research Board. 1981. "Concrete Block Test PavementInstalled at ARRB," Australian Road Research, Vol 11, No. 1, Vermont,Australia.
Barber, S. D., and Knapton, J. 1980. "An Experimental Investigationof the Behavior of a Concrete Block Pavement with a Sand Sub-Base,"Proceedings, Institute of Civil Enineers, Part 2, Vol 69, 1 March,pp 139-154.
Besser Company. (No date). "The Besser Company Paving Stone Manual,"Alpena, Mich.
, Brown, D. H., and Ahlvin, R. G. 1961. "R vDesign for Flexible Highway Pavements at Military Installations," Tech-nical Report No. 3-582, U. S. Army Engineer Waterways Experiment Sta-tion, Vicksburg, Miss.Cedergren, H. R. 1974. Drainage of Highway and Airfield Pavements,John Wiley and Sons, New VOk.
I Cement and Concrete Association. 1976. "Continental Experience ShowsGrowth Potential for Interlocking Paving," Precast Concrete, Vol 7,No. 6. United Kingdom, pp 291-298.
Cement and Concrete Association, 1978. "Concrete Block Paving, Model. " ,. . . Specifications Subject to Adaption," Wexhaw Springs, United.Kitgdos.
Clark, A. J. 1978Pavi Research and Developent," Concrete,Vol 12, No. 7, Cement and Concrete Association, United Kingdom,pp 24-25.
Clark, A, J. 1919. '"ater Penetration Through Newly Laid Coocrete I :Block Paving," Technical Report 529, Cement aud Concrete Association,.4 Wexham Springs, United Kingdom.
Clack, A. J. 1980. "Freeze-Thaw Durability Tests Upon Concrete PavingBlock Specimens," First International Conference on Concrete BlockPaving, University of Newcastle-Upon-Tyne, England.
Cruicksh nk, J. W. 196. "Interia Guide'to the Dtsig and construction]. :of Interlotkio, Sett Pavesent=,"4 Technical Note 32, Cement and ConcreteV Association, Australia4. oe Cmn
Dawson, A. J. 1980. "Mix Design for Concrete Block Paving," First In-ternational Conference on Concrete Block Paving, University of Newcastle-Upon-Tyne, England.
Department of the Army. 1965. "Pavement Designs for Frost Conditions,"Technical Manual 5-818-2, Washington, D. C.
Department of the Army. 1971. "Flexible Pavements for Roads, Streets,Walks and Open Storage Areas," TM 5-822-5/AFH 88-7, Chapter 3, Washing-ton, D. C.
Department of the Army. 1975. "Standard Practice for Concrete Pave-ments," TH 5-822-7/AFN 88-6, Chapter 8, Washington, D. C.
Department of the Army. 1979. "Use of Precast Concrete Block Pave-ments," ETL 1110-3-310, Office, Chief of Engineers, Washington, D. C.
Department of the Army. 1980. "Engineering and Design, Flexible Pave-ments for Roads, Streets, Walks, and Open Storage Areas," TM 5-827-5/AFI 88-7, Chapter 3, Washington, D. C.Department of Defense. 1978. "Flexible Pavement Design for Airfields,"TM 5-825-2/UN! 21.3/AFH 88-6, Chapter 2, Washington, D. C.Dreijer, P. A. 1980. "Laboratory and Fieldwork on Block Paving,"First International Conference on Concrete Block Paving, University of .Newcastle-Upon-Tyne, England.Federal 11ighway Aduini st ration. 1980. "iighway Subdrainago Design,"Report No. FRWA-75-80-224, Washington, 0. C.Gerrard, J. S. 1980. "Review of Block Paving in United Kingdom Ports,"First International Conference on Concrete Block Pavements, University 4of N castle-Upon-Tyne, England. t U o ne P Bc
Harris, 0. 1978. "An Examination of the Use of Concrete Paving Blocksas a Pavement Surfacing Course," Report No. 76/22, haiu Roads epart"utof Western Australia, Materials Eogineering Divisiou, Australia. JHodgkirson, J. A., and Horrish, C. F. 1980. "lterim Guide to the De-sign of hnterlocking Concrote Pavements for Vehicular Traffic," TN 34,Cemenst and Concrete Association. Australia,
ellersawnn, G. 11, 1980. "Urban Block Plavig in the Netherlands," • . .1First loterlntioull Conference on Concrete Block Pavements, University - . .
, 1 of Necastle-lpon-tye, England.
KRapton, J. 1016 . "The Vesigo of Cojcrete Block Roads," Report"" No. 42.515, cemt aud Coucrete Asociatiou, W/exhkm'Spriugs, United
Kingdom.
Knapton, J., and Barber, S.0. 1979. "The Behavior of a Concrete Block* ].Pavement," Proceedings., Iastitute of Civil Engineers, Part 1, Vol 66. f
Knapton, J., and Barber, S, U. 1980. "U. K. Research into ConcreteBlock Pavement Design," First Intentational Conference on ConcreteBlock Paving, Uaivetaity of 1Necastle-U1poo-Tyne, England. j
. i118 .
.-... , -- ,":" '.- , :'' -~ ; : ,4 ."
I m ...... . -... _. . .. -...... -•-
Knapton, J., and Lilley, A. A. 1975. "A Study Visit to Germany and Den-mark," Cement and Concrete Association, Wexham Springs, United Kingdom.
Kuthe, Ernst-Otto. 1980. "Concrete Paving Block Today," First Inter-national Conference on Concrete Block Paving, University of Newcastle-Upon-Tyne, England.
Lane, R. 0. 1978. "Abrasion Resistance," Chapter 22, Significance ofTests and Properties of Concrete and Concrete-Making Materials, ASTMSpecial Technical Publication 69B, American Society for Testing and
Materials, Philadelphia, Pa.
Lesko, S. 1980. "The Use of Concrete Block Pavements for Highways,"
First International Conference on Concrete Block Paving, University ofNewcastle-Upon-Tyne, England.
Litvan, G. and Sereda, P. 1978. "Particulate Admixtures for EnhancedFreeze-Thaw Resistance of Concrete," Cement and Concrete Research,Vol 18, No. 1, pp 53-60.
Lilley, A. A. 1980. "A Review of Concrete Block Paving in the U. K.4Over the Last Five Years," First International Conference on ConcreteBlock Paving, University of Newcastle-Upon-Tyne, England.
Lilley, A. A., and Clark, A. J. 1978. "Concrete Block Paving forLightly Trafficked Roads and Paved Streets," Report No. 46.024, Cementand Concrete Association, Wexham Springs, United Kingdom.
Lilley, A. A., and Collins, J. R. 1976. "Laying Concrete Block Paving,"Report No. 46.022, Cement and Concrete Association, Wexham Springs,United Kingdom.
Lilley, A. A., and Walker, B. J. 1978. "Concrete Block Paving forHeavily Trafficked Roads and Paved Areas," Technical Report 46.023,
Cement and Concrete Association, Wexham Springs, United Kingdom.
Mather, B. 1975. "Durability of Concrete Construction--50 Years ofProgress," Vol 101, No. C01, Journal of the Construction Division,American Society of Civil Engineers, New York, pp 5-14.
Meyer, A. 1980. "Materials and Specifications in West Germany,"First International Conference on Concrete Block Paving, University ofNewcastle-Upon-Tyne, England.
Miller-Cook, D. 1980. "Concrete Block Paving - Survey of IndustrialApplications in the U. K.," First International Conference on ConcreteBlock Pavements, University of Newcastle-Upon-Tyne, England.
Morrish, C. F. 1980. "Interlocking Concrete Paving - The-State-of-the-Art in Australia," First International Conference on Concrete BlockPaving, University of Newcastle-Upon-Tyne) England.National Concrete Masonry Association. 17."ConcreteBlcPaentDesign and Construction," Herndon, Va.
National Concrete Masonry Association. 1980. "Concrete Block Pavements
Structural Design," TEX 115, Herndon, Va.
119
....... I
Nettles, E. H., and Calhoun, C. C. 1967. "Drainage Characteristics ofBase Course Materials Laboratory Investigations," Technical ReportNo. 3-786, U. S. Army Engineer Waterways Experiment Station, Vicksburg,Miss.
Neville, A. M. 1973. Properties of Concrete, 2d ed., John Wiley andSons, New York.
Ozyildirim, C., and Sprinkel, M. 1982. "Durability of Concrete Contain-ing Hollow Plastic Microspheres," MS 5161, American Concrete Institute.
Patterson, W. D. 0. 1976. "Functional Pavement Design for ContainerTerminals," Proceedings, The Eighth Australian Road Research Board Con-ference, Vol 8, Part 4, Perth, Australia, pp 22-29.
Pesch, L. 1980. "Concrete Paving Blorks in Industrial Construction in "" .* West Germany," First International Conference on Concrete Block Paving,* University of Newcastle-Upon-Tyne, England.
Powers, T. C. 1975. "Freezing Effects in Concrete,'" SP 47-1, Durabil-ity of Concrete, American Concrete Institute, Detroit, Mich.
Precast Concrete. 1979. "Uniblocks at Port Rashid," Vol 10, No. 12(Dec), United Kingdom, pp 5-84.
Seddon, P. A. 1980. "The Behavior of Concrete Block Paving UnderRepetitive Loading," First International Conference on Concrete Block
Shackel, B. 1978. "An Experimental Investigation of the Response of
Interlocking Concrete Block Pavements to Simulated Traffic Loading,"LSeminar on Interlocking Concrete Block Pavements, Australian Road :WResearch Board, Australia.
Shackel, B. 1979. "A Pilot Study o h efrac fBokPvnUnder Traffic Using a Heavy Vehicle Simulator," Proceedings, Symposium
on eaaL,ConcreePaving Blocks, Johannesburg, South Africa.
Shackel, B. 1980a. "The Performance of Interlocking Block PaeetUnder Accelerated Trafficking," First International Conference 'on Con- p .
crete Block Paving, University of Newcastle-Upon- Iynie, England.
Shackel, B. 1980b. "The Design of Interlocking Concrete Block Pave-ments for Road Traffic," First Internatioool Conference on ConcreteBlock Paving, University of Newcastle-Upon-Tyne, Eugland..
Shakel B, ad Aor, I . 1978a. "The Evaluation of InterlockingBlock Pavements,"~ An Interim.Report. Proceedings, Annual Conference, -
Concrete Masonry Associatiop, Australia. V-
Shackel, B., and Arora,. . 1 918b. "The Application of a Full-ScaleRoad Simulator to the Study of Highway Pavements," Ausralian RoadResearch, Vol 8, No.* 2, Juno.
Sharp, K. B. 1980. "An InitialStudy into Concrete block Paving," -
First International Conference on Concrete Block Paving,-University-of
Newcastle-Upon-Tyaq, England,.
120
Spears, R. E. 1978. "Concrete Floors on Ground," Portland Cement As-sociation, Skokie, Ill.
Van der Vlist, A. A. 1980. "The Development of Concrete Paving Blocksin the Netherlands," First International Conference on Concrete BlockPaving, University of Newcastle-Upon-Tyne, England.
Van Leeuwen, H. 1980. "Use of Block Paving in European Ports," FirstInternational Conference on Concrete Block Paving, University ofNewcastle-Upon-Tyne, England.
Von Szadkowski, G. 1980. "The Effect of Pigments on the Quality of -
Concrete Blocks," First International Conference on Concrete BlockPaving, University of Newcastle-lUpon-Tyne, England.
Wiley, C. C. 1919. "MHodern Pavements (A Discussion on Various Types ofPavementE, Their Relative Merits and Comparisons with Concrete Roads),"Brick and-ClaX R'..cord.
Working Commnittee on Concrete Block Paving. 1965. "Code of Practicefor the Construction of Concrete Block Paving," Forschangsgellschaftfur das Strassenwesen (Road Research Association), Germany.
Yoder, E. J., and Wand M. W. 1975. Principles of Pavement Design,
121
5
- - - -- - - .- - - - .. - ~ ~ -- - ---------
Table 1
West German Use of Paving Block in 1972*
Roads 36.4 percentIndustrial Areas 29.1 percentPrivate Drives 12.5 percentPedestrian Ways 12.0 percentParks, Schools, etc. 6.8 percent
Other 3.2 percent
*From Cement and Concrete Association (1976).
Table 2
CoMparison of Paving Block Strength Requiremnts
Required
U. S. National Concrete Tp fTs tegh s
Masonry AssociationArea subject to
freeze-thaw cycles Compressive 8000
Area not subject to- freeze-thaw cycles Compressive 6000
United Kingdom Cement andConcrete Association Compressive 7250
Netherlands NEN 7000 Flexural 8551-German DIN 18501 Compressive 8700
J Table 3EsttmIated Concrete Block Pavement Construction Outg!t
Estimated Output*1 Source yd /man fat.Lilley and Clark (1978) 30 to 60D. G. Frandsen (1978)* .20 to 30!4orrish (1980) 2.Kelleremann (1980) 59.8 to 167-
*Personal communication, 1978, Hr. D. G. Franidsen,U. S. Army Engineer Division, Europe$ Vraukfurt$Germany.
4
J4J
4 0 41 41 4J 41
4 00
ON~ 4) 0 4 II
'.44
4)4444 4
P4
14, 0 40l
00
to0 u~w 14r~ 4~W 4 4)L
$4.
'0 4) 4) o#
4) 44,41.acu
El-A 2 1HV S1 -
m W 0.
44 .2 0 00 e
41 14 o
1- a)..
U--1
Table 5Cost for Paving Block in Place
Project2Year Location Size, ft Cost, $/ft2 Source
1976 Fulda Barracks, 29,900 1.14 Personal comunicationi,Germany Mr. Frandsen, Corps
* of Engineers
1977 El Cerrito, 9,280 3.00-3.50 Personal communication,*Calif. LTC Delano, Corps of
Engineers
1978 Perth, Australia -- 1.24 Hiarris (1978)
Dover, United 290,000 1.17-1.54 Gerrard (1980)Kingdom
1979 Eastern U. S. -- 1.85-3.00 Personal commnunication
California -- 3.00-5.50 Personal. comunication
1980 Netherlands -- 1.39 Van Leeuwen (1980)
Table 6
* Nominal Currency-Exchangte Rates
Excang RteIi.S.DollarsNation Cure 1976 1977 1978 1979
Australia Dollar -- 1.10 1.1565 1.1295
Netherlands Guilder -- 0.408 0.4592 0.4982
*United Kingdom Pound -- 1.72 1.9325 2.261
* **West Germay Mlark 0.4265 0.424 0.4984 0.5468
ii ______ I *.ii21.i 2A7
Table 7
Variation in 1978 Bids, Dover, U. K. (After Gerrard (1980))
Cost for 9.8-in. Cost forLean Concrete 80 mm BlockToa s
Bidder No. ____________ $/ft 2$/ft
1 0.78 1.17 1.952 1.44 1.54 2.98
3 1.10 1.48 2.584 1.00 1.22 2.22
5 0.85 1.54 2.39I6 0.91 1.44 2.35
Mean 1.01 1.40 2.41
Coefficient ___________________________of____
Variation** 0.24 0.12 0.14
*Dollars are U. S. dollars.ft Coefficient of variation =Standard deviation + mean.
Si Table 84
Initial Construction Costs for CoM arable
Pavements, U. S. Dollars/ft2
Appli- SprayCountry Year cation Block Asphalt Seal Concrete Reference
Australia 1978 Road 1.24 0.28-0.41 0.23 -- Harris('4978)
United 1980 Port 4.28 4.36 -- 4.85 GerrardKingdom (1980)
Netherlands 1980 Port 2.55 2.55 -- 3.70 Van Leeuwen(1980)
Table 9
Trffcamaeto BlockRoads
Annual-map On Marshy Soil. On Sandyil
Depressions in ft2 per mile of road 107.6 64.6v
Broken blocks in ft2 per mile of road 10.8 10.8
NOTE; One mile of road averaged 32,300 ft2 of pavement.
7.7
* . ~ 4,
Gi 41 a a
14 41 41 41 ~ .4 4
W04 .40
~4~ .J ~ + 10 U 01
0 0 0 0 ci 0a. 0- P- 0 v4
4 $4 04 $4 0. 40 >1)
41 44 ~4.444 t 4.41- ca po.
) 000 14-4
4. H~'~0.
di4 F4 i-I*'. N. z, '. ~ . 0V '.
w-4 4-40o o
P-4 a)0)0 .- 4 4 .4
44 to 4100~' V)4-0 ?AW 4
0V w.@ ~ +1
~r 0 ~.r 0 ~ 0 .4* ,4
Q A j )0 E %
3 I40 04 0 01
N r-I
, wl U34
-----.- - ----- _..._._...
Table 11Results of Water Penetration Tests (Clark 1979)
Surface WaterPaedAra Water Flow, % Penetration, Length
TetSlope Application of Applied % of Applied of TestTest_ ft2 Rate, in./hr Water Water* moin
1 2.5 32.3 1.81 76 20 60
2** 2.5 32.3 1.77 80 16- 60
3 1.0 32.3 2.09 78 18 60
4 1.0 32.3 0.87 70 24 70
5 1.0 32.3 2.05 81 17 45
6 1.0 34.9 1.02 79 21 90
7 1.0 34.9 1.81 80 18 40
8t 1.0 34.9 1.06 77 13 65
9t 1.0 34.9 1.54 83 10 601011 1.0 34.9 1.57 89 1 48
lit 1.0 32.3 1.81 81 15 71
r
*As determined from measurement at the drainage outlet..Repeat of test 1 conducted 28 days later.
t 20 percent clay dust toixced vith sand in jloints.tt Same pav~ement ed for tests 8, 9, and 10. Test 10 run 24 hr aftertests 8 and 9.
$Top soil brushed into surface of joints after construction.
Table 12
Industrial Road Test Section Design (Barber and Knapton 1980)
Thicknessof Base
ItmSubgrade Course, in. Type of Block
I Sandy clay 11.8 Rectangular
2 Heavy clay 15.7 Rectangular
3 Sandy clay 19.7 Rectangular
4 Sandy clay 23.6 Rectangular
5 Sandy clay 27.6 Rectangular
6 Sandy clay 27.6 Shaped
7 Sandy clay 23.6 Shaped
8 Sandy clay backfill 19.7 Shaped
9 Sandy clay backfill 15.7 Shaped
10 Sandy clay backfill 11.8 Shaped
Table 13
Industrial Road Subgrade Properties (Barber and Knapt.on 1980)
Natural Backfill Heavy ClayISubgrade Sbi.rde Subgrade
Applicable test items 1, 3-7 8-10 2
Soil type Saudy clay Sandy clay Heavy clay
Casagrande classification CL CL CL
Plasticity index, percent 15.5 13.5 36.4Moisture content, percent 12.9 12.3 14.3
CBR, percent 5 3* 2
.1General quality Fair to Poor Poor Poor.
. .I*Remolding upon compaction reduced this to 2.percent.
I.& 3
Table 14
Industrial Road Traffic Deformations (Barber and Knapton 1980)
BaseThickness Traffic Deformation, in.
Item in. TO0* 750* 1200* 1500* 1750* 230 4070*
1 11.8 0.65 0.63 0.77 0.81 0.83 0.87 0.89
2 15.7 1.08 1.12 1.18 1.18 1.22 1.26 1.32
3 19.7 0.61 0.59 0.71 0.71 0.73 0.83 0.87
4 23.6 0.71 0.73 0.73 0.75 0.79 0.85 0.89
5 27:.6 0.43 0.47 0.51 0.55 0.61 0.65 0.69
6** 27.6 0.49 -0.49 0.55 0.55 0.57 0.59 0.61
7* 23.6 0.61 0.65 0.67 0.67 0.73 0.73 0.79j
8 19.7 0.67 0.77 0.81 0.83 0.83 0.87 0.91
9* 15.7 1.93 2.05 2.09 2.09 2.11 2.12 0.47t
10** 11.8;' 3.66 3.98 4.04 4.07 4.13 4.17 0.711t
Traffic in equivalent 18-kip axles.
**Shaped blocks '; all others rectangular.1Item relaid after.2320 equivalent 18-kip axles.
Table 15
:Subj~ade and Base Material Properties (Shackel 1979)
Sugrde Base
Liquid limit, percent 3.
Plastic limit, peiteat 1.
Plasticity index, percent, 18.0 7.9
1WodifiO4 AASHTO optimum dry density, lb/ft3 119.7 124.5
Miodified MASITO optimum moist coutent, percent 13.0 9.3IIn plate dry- denaiItyo 109.4187
Pe~geat of ucidified AASHTO dry diensity .91.4 95.3
0.10 40 000
'o 1m 00 COD 00r 0 O In V-1 lOS In NO
040
u 0a O 00 000-OTcd 0O 50 10 1 1 0 I
S4 -
Ch* W
00 N C*4
rn0
t~~~ Q Sm I I i I I L 1 1 in 1 n
U, 0 1 1
-? --
Table 17
WES Block Properties
RetangularProperty Test Procedure Block Z Block Unistone
Absorption, percent ASKi4 C 67-78
5 hr 3.32 2.28 2.60
24 hr 3.93 3.25 3.94
5 hr boiling 6.97 6.11 7.64
Saturation coeficient 56.09 53.18 51.62 ISandblast abrasion
cofiin /c . CRD-C 58-78 1.57 2.22 1.72
Freeze-thaw ASf C 67-78 -Negligible results-
Density, lb/ft3 CRD-C 7-79 139.3 143.8 133.7 *
Flexural strength, psi ASIh C 67-78 1,152 1090 1004
Spitting tensile, psi CRlD-C 77-72 .1,408 1170 .1438
Compressive strength, psi ASI C 67-78 10,078 8152 8622 1
V ..
, 4.
Vn
I, .. ,-
t * .
- .. ,,~. •
.5.4 C44 a
'44x
0) 0A I 0P mi r, C40 %0 en:04CU Q
0 Col 0 LM %0O
-T m00-2c o m
"m %0 IC" I
as 04 1
0 a4
CM Vo OM , ~ 0
44 pt
04
wt 0 W
U0
.; ;INK
*..4 N N004.4 y4l
.w 4-4 004 a~ r O C9
* E-4 Ch00 -I
4)0 0)4-u 41)
*4 m
44 -i 0 0 0 ~ O -
CN $4 .
0 44O
0 44 4 N .- 4 r4
'44 ifn NO4 L~ Q ~ 41%
0. H1 p4 0 00 ~4J
411 '-4 43
4i -it H- c 00C > 4
i- i4 ,- 41 I U I a )
H sI 1. 4.4 No I N I '. I LI) 1 0*w 4-W
I..u 0 4.)1$ 4~.
44434 M i CAU
0) 0c N. 000 C'0Wrh
4 $4 r -I N 41 04
ci~~~~~- 4. 0 O-4- 3U0i
v co '.0 0. -H ' 4JHri .4*i 4 014 -h M 4)i4-1 00 4 00 000 00 mi
0 44 O 4 -1~
4.) C; ci $4(a 0n 0" 0 30Q4. 4*.4-4 r4.)04.i.l
~0 0 0 U34 4
0 C) 0 0 0 C04 4H ) '
41 4.1 P- 4.1 U. 0
4~ 4J4. 4J J 4. .
41 o4 W 04343 U
.F4 H 41 .0
.4,-.~1 tI\ 14-
4- .. V
t"4 J 4 44
Table 20
Comparison of Predicted and Actual
Test Section Performantce
Ite CBR Dept* Predicted** Equivalentt Predictedtf Oerations
No. Layer in. Operations Depth, in. Operations. at Failure
991 Base 100 4.1 >10 6.5 >0>8017t
Clay 7 8.1 3,400 10.5 45,000crust
Clay 4 14.1 22,000 16.5 110,000
Base 100 4.1 >10~ >. 10 ~ >8071*
Subbase 20 8.1 2 x 108 10.5 >10~
Clay 3 .12.1 1,150 14.5 5,000
3 .Base 100 3.4 >10~ .
Subbase 20 7.4 2.4 x 10 ~ 10.5 >10 ~ 5808
Clay 3 11.4 3,000 14.5 5,000
*Actual depth to surface of layer.**Using actual depth and CDR curves in~ Figure 78.
6.s in. (Dprwtof the Arey 1979).j * These test atectog did not fail.
lomN1 NO ..
-------------
* In accordance with letter from DAEN-RDC, DAEN-ASI dated22 July 1977, Subject: Facsimile Catalog Cards forLaboratory Technical Publications, a facsimile catalog
* i card in Library of Congress MARC format is reproducedbelow.
Rollings, Raymond S.
Concrete block pavements Iby Raymond S. Roilings(Geotechnical Laboratory, U.S. Army Engineer WaterwaysExperiment Station). -- Vicksburg, Miss. : The StationSpringfield, Va. ; available f'rom NTIS, 1983.
121, [131 p. :ill. ;27 cm. -- (Technical reportGL-83-3)Cover title."March 1983."1Final report.
*"Prepared for Office, Chief of Engineers, U.S. Armyunder Project No. 4A1611O2AT22, Task Area AO, WorkUnit 005."
Bibliography: p. 117-121.
1. Concrete blocks. 2. Pavements, Concrete.1. United States. Army. Corps of Engineers. Officeof the Chief of Engineers. 11. U.S. Army Engineer
Rollings, Raymond S. (ad2
Concrete block pavemnents I by Raymond S. Rollings :..1985.
Waterways Uporiment Station. Geotechnical Laboratory.IIll. Title IV. Seriest Technical report (U.S. ArisyEngineer Waterways flxpoa'inmet Station) SL-83-3.TA7.WS4 iwo.GL-63-3
mI
_____________________________________________a
